front-of-pack nutrition rating systems and symbols … · front-of-pack nutrition rating systems...
TRANSCRIPT
Front-of-Pack Nutrition Rating Systems and Symbols in Canada
by
Teri Elyse Emrich RD, MPH
A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy
Department of Nutritional Sciences University of Toronto
© Copyright by Teri Emrich 2014
ii
Front-of-Pack Nutrition Rating Systems and Symbols in Canada
Teri Elyse Emrich Rd, MPH
Doctor of Philosophy
Department of Nutritional Sciences University of Toronto
2014
Abstract
Front-of-pack (FOP) nutrition rating systems and symbols provide simplified information on the
nutritional content of foods. Despite concerns that the presence of multiple FOP systems may be
confusing and misleading to consumers, the government has rejected recommendations to
implement a single, standardized FOP system claiming that the current unregulated marketplace
is already meeting the needs of Canadians. The overall objective of this work was to generate
Canadian evidence to determine the legitimacy of the government’s position.
The specific objectives were to evaluate the extent to which foods meeting the criteria of
voluntary FOP systems are identified by the system’s symbol, to compare the nutritional quality
of foods with and without FOP systems, to describe Canadians’ attitudes towards FOP systems,
and to determine experimentally the most liked and understood FOP system by consumers. We
generated evidence through three studies.
Studies 1-2 employed a 2010 database of Canadian packaged food labels. Compared to products
without FOPS, products with FOPS were not being used to market foods consistently lower in
calories, saturated fat, sodium, and sugar in 10 food categories and 60 subcategories.
Significantly many more products qualified to carry two major Canadian FOPS than actually
displayed these FOPS on their label.
Study 3 used a national survey panel to test Canadians preferences for four FOPS. 86% of
consumers supported the implementation of a single, standardized FOPS. Nutrient-specific
systems (a traffic light and a Nutrition Facts table based FOPS) were preferred and rated more
consumer friendly than other systems.
iii
In summary, FOPS are not being used to promote the healthiest products in the Canadian
marketplace, and could be misleading consumers. Despite the government’s position not to
adopt a single, standardized FOPS, Canadian consumers would support such an initiative and
would prefer a nutrient specific FOPS.
iv
Acknowledgments
I would like to thank my supervisor, Dr. Mary L’Abbe, for four years of continued support and
mentorship. Dr. L’Abbe continually encouraged me to take advantage of whatever opportunities
came my way, be they academic, professional, or personal, and was instrumental in opening the
doors that led to so many of those opportunities. Over the course of my PhD, I have worked
harder and learned more than during any other period of my life, but Dr. L’Abbe always
reminded me to make time for fun and to celebrate every accomplishment – no matter how
minor.
My thesis committee members, Dr. Joanna Cohen, Dr. Marco DiBuono, and Dr. Wendy Lou,
have always been available to provide thoughtful input into my work, bringing in fresh
perspectives that made the research much richer. I am grateful to these three people for investing
their valuable time and energy in my development as a professional and a researcher. I am also
to grateful to Dr. David Jenkins and Dr. Sara Kirk who served as my external examiners for my
Departmental and Senate defenses, respectively. They both challenged me in ways that made me
see my work from a new perspective. Laura Pasut has been my professional mentor for the later
years of my PhD. Her insights have opened my eyes to the possibilities ahead of me and have
helped to focus my sights on life after my PhD.
In the last four years I have been so fortunate to work with a wonderful and diverse group of
fellow graduate students and trainees who were readily available to lend a helping hand
whenever needed. Dr. Joanne Arcand, has made a significant contribution to the training of all
the students in Dr. L’Abbe’s research group, myself included. I have learned so much from her
about the discipline and dedication required of academic research. I am also grateful for the
contribution of Alyssa Schermel to the development of the Food Label Information Program,
without which much of my research would not have been possible. Christina Wong and Mary
Scourboutakos have been my constant companions on this journey. They commiserated with
me, provided listening ears, and made me feel that I was never alone in this. Ying Qi, of Dr.
Lou’s research group, made a significant contribution my work and exhibited endless patience
when working with me on my analysis, for which I am extremely grateful. Julio Mendoza was
instrumental in my work with the Canadian Consumer Monitor, and his management of this
consumer panel facilitated my survey work.
v
My family was my biggest cheerleader over the course of my graduate program and encouraged
me to persevere and to keep moving forward. I thank my parents, Margo and Walter, my sisters
Katie and Andi, and my nephews Jack and Henry – they were my escape from my academic
world and have brought my life tremendous joy. My grandparents, Jack and Anne Disher, I have
never felt that anyone is more proud of my accomplishments and me than these two. A phone
call with them could help restore my confidence in myself even on my toughest day.
Finally, to my wonderful partner in life Aaron, you have been my biggest supporter on this
journey. My pursuit of this dream meant that many aspects of our life together were put on hold,
but you never complained and always saw the bright side. You bore the brunt of my stress and
anxiety and were there to talk me down whenever I needed you. I am so happy that you came
through this with me and I am excited to finally begin the next chapter of our life together.
I have been fortunate to receive support for my research and training from several organizations.
The Canadian Institutes of Health Research provided funding through a Banting and Best
Doctoral Scholarship, a Population Intervention for Chronic Disease Prevention: A Pan-
Canadian Program Training Fellowship, and a Public Health Policy Training Fellowship. In
particular, the Population Intervention for Chronic Disease Prevention contributed to my
research and training far beyond the funding provided with its most significant contributions
coming through the training and fellowship opportunities it provided. The Advanced Foods and
Materials Network also provided funding early in my training.
vi
Table of Contents
Acknowledgments iv
Table of Contents vi
List of Tables ix
List of Figures x
List of Abbreviations xi
List of Appendices xii
Chapter 1 1
1 Introduction 1
Chapter 2 3
2 Background and literature review 3
2.1 Diet related chronic disease 3
2.2 An overview of nutrition information on food labels 4
2.2.1 Nutrition labelling 4
2.2.2 Nutrition marketing 6
2.3 An overview of FOP nutrition rating systems and symbols 11
2.3.1 History of FOP symbols 12
2.3.2 FOP nutrition rating system and symbol categories 13
2.3.3 Nutrient profiling 17
2.4 FOP systems and the nutritional quality of foods 20
2.4.1 FOP systems and reformulation 20
2.4.2 The nutrient profiles of foods with and without FOP systems 22
2.5 Influence of FOP systems on perceptions of nutritional quality 24
2.6 Is there a need for a single, standardized FOP system? 27
2.6.1 The ideal FOP system for use with consumers 28
2.7 Summary 35
Chapter 3 36
3 Scope and Hypotheses of Thesis 36
3.1 Scope and Objectives 36
vii
3.2 Specific hypotheses 36
3.3 Preview of Chapters 4-6 37
Chapter 4 39
4 Food products qualifying for and carrying front-of-pack symbols: a cross-
sectional study examining a manufacturer led and a non-profit organization led
program 39
4.1 Abstract 40
4.2 Background 42
4.3 Methods 45
4.4 Results 50
4.5 Discussion 54
4.6 Conclusions 57
Chapter 5 59
5 Front-of-pack symbols are not a reliable indicator of products with healthier
nutrient profiles 59
5.1 Abstract 60
5.2 Introduction 61
5.3 Methods 65
5.3.1 Data analysis 66
5.4 Results 66
5.4.1 Differences between products with and without FOP symbols by food category 67
5.4.2 Differences between products with and without FOP symbols by food subcategory 67
5.4.3 Comparison of different FOP symbol types 68
5.5 Discussion 72
5.5.1 Limitations 73
5.5.2 Conclusions 74
Chapter 6 75
6 Consumer perceptions of the Nutrition Facts table and Front-of-Pack nutrition
rating systems 75
6.1 Abstract 76
6.2 Introduction 77
6.3 Materials and Methods 78
viii
6.3.1 Participants 78
6.3.2 Survey Design and Procedure 79
6.3.3 Statistical Analyses 80
6.4 Results 82
6.4.1 Participants 82
6.4.2 Consumer friendliness of FOP systems 82
6.4.3 Effect of NFt on consumer friendliness of FOPs 82
6.4.4 Conceptual Understanding 82
6.4.5 Effect of FOP on perceived overall healthiness and nutrient content 83
6.4.6 Consumer preferences for FOP systems 84
6.5 Discussion 93
6.5.1 Conclusion 95
Chapter 7 96
7 Discussion 96
7.1 General Discussion 96
7.1.1 Evaluation of existing FOP system policy 97
7.1.2 Evaluation of potential FOP system policy 100
7.2 Future directions 101
7.2.1 Policy 101
7.2.2 Research 109
Chapter 8 112
8 Summary and Conclusions 112
References 114
Appendices 128
Appendix A 129
Appendix B 134
Appendix C 136
Copyright Acknowledgements 145
ix
List of Tables
CHAPTER 4
Table 4-1 Nutrient criteria for crackers to qualify for Health Check™ and Sensible Solutions™ symbols
CHAPTER 5
Table 5-1 Comparison of the nutrient content per reference amount of products with and without front-of-pack nutrition rating symbols*
CHAPTER 6
Table 6-1 Demographic characteristics of survey respondents (n=3029)
Table 6-2 Mean scores of consumer friendliness by FOP system and NFt
Table 6-3 Mean scores of statements describing the NFt and FOP systems
Table 6-4 Mean healthiness, calorie and nutrient content scores of a frozen meal and breakfast cereal by NFt and FOP
CHAPTER 7
Table 7-1 US Food and Drug Administration and US Department of Agriculture nutrient profiling criteria for “healthy” claims
x
List of Figures
CHAPTER 2
Figure 2-1 Nutrition information on prepackaged foods in Canada
Figure 2-2 Examples of different front-of-pack symbol types
Figure 2-3 Conceptual model of FOP system use
CHAPTER 4
Figure 4-1 Front-of-pack symbols evaluated in the present study
Figure 4-2 Proportion of food products that qualified for, compared with the proportion of food products that carried, the different front-of-pack symbols
CHAPTER 5
Figure 5-1 Examples of different front-of-pack symbol types
Figure 5-2 Difference in calorie, saturated fat, sodium, and sugar content per reference amount between products with and without front-of-pack symbols by symbol type
CHAPTER 6
Figure 6-1 Front-of-Pack Nutrition Rating Systems evaluated in the randomized mock package experiment.
xi
List of Abbreviations
ANCOVA Analysis of Covariance
CCM Canadian Consumer Monitor
FLIP Food Label Information Program
FOP Front-of-Pack
NFt Nutrition Facts table
US United States
xii
List of Appendices
Appendix A Proportion of food products that qualified for Health Check™ compared to the proportion of food products that carried the system's symbol by subcategory (N=7503)
Appendix B Proportion of food products that qualified for Sensible Solutions™ compared to the proportion of food products that carried the systems symbol by subcategory (N=3009)
Appendix C Comparison of the nutrient content per reference amount of products with and without front-of-pack nutrition rating symbol nutrition marketing by product subcategory
1
Chapter 1
1 Introduction
Globally, diet-related chronic disease – including obesity, diabetes, cardiovascular disease,
cancer, osteoporosis, and dental diseases – are on the rise (1). Within Canada, three of these
conditions – cardiovascular disease, cancer, and type 2 diabetes – are among the leading causes
of death (2). In order to prevent these diet-related chronic diseases, the World Health
Organization recommends that healthy populations limit their intake of saturated fat, trans fat,
cholesterol, sugar, sodium, and total energy while ensuring adequate intakes of fat,
polyunsaturated fat, carbohydrate, protein, fibre, and vitamins and minerals (1,3). Similarly,
2007’s Eating Well with Canada’s Food Guide recommends Canadians limit foods and
beverages high in calories, fat, sugar or salt, as well as trans fat while meeting their needs for
vitamins, minerals and other nutrients (4).
In order to make healthy food choices, in line with the above-mentioned recommendations, the
WHO suggests that “consumers require accurate, standardized and comprehensible information
on the content of food items” (3). To that end, Canada has been actively engaged in providing
consumers with information on the nutritional content of foods through regulated nutrition
labelling since 2003 (5). However, research is beginning to show that the nutrition labelling
regulations have not been entirely effective in helping Canadians make healthier choices. While
the majority of consumers report reading food product labels at least sometimes, actual use of
food labels while shopping has been found to be much lower than reported in international
studies (6-9). In addition, while consumers commonly report understanding nutrition labels,
actual understanding is more moderate than reported (7-11). Many Canadian consumers are
confused about the technical and numerical portions of nutrition labelling, while others struggle
to put this information in the context of a healthy diet (12). In some cases, food label
understanding is hampered by consumers’ lack of knowledge of nutrient requirements and
reference amounts (10). Consumer confusion and uncertainty about one or more elements of the
nutrition label impedes their ability to use the label to guide their food choices. Indeed, experts
have raised concerns that the current labels are too complex and require too much time to
interpret (13). As a result, the House of Commons Standing Committee on Health, as well as
government working groups, have called for the development and implementation of a simplified
2
front-of-pack (FOP) nutrition rating system or symbol (13,14). Internationally, several
governments have already begun to pursue such a strategy (15). Both the Australian and United
Kingdom governments have introduced standardized, voluntary FOP systems with government-
developed nutrient profiling criteria (16,17). Australia has developed a health star FOP system
while the United Kingdom has developed a traffic light FOP system.
In contrast, the current lack of specific regulations for FOP nutrition rating systems in Canada
has led to the proliferation of a number of different FOP systems, each with their own criteria
(13,18,19). As a result, concern has been raised that the proliferation of competing FOP systems
may lead to greater confusion and mistrust among consumers trying to choose food products on
the basis of the nutrition information available on food packages (13,18-22). Furthermore,
existing FOP systems have been the subject of criticism for using nutrition criteria that aren’t
stringent enough to exclude unhealthy choices, and as a result, do not guide consumers’ choices
towards truly healthy products. In light of calls for the development and implementation of a
simplified FOP nutrition rating system for Canada, the potential contribution of FOP systems to
supporting consumers’ ability to identify healthier choices should be studied. The overall
purpose of this research is to explore the healthfulness of foods designated by FOP nutrition
rating systems in Canada and the consumer friendliness of different FOP symbols.
3
Chapter 2
2 Background and literature review
2.1 Diet related chronic disease
Worldwide, chronic non-communicable diseases are the leading cause of death, and deaths due
to chronic disease are on the rise (23). Four chronic diseases – cardiovascular diseases, diabetes,
cancer, and chronic respiratory diseases – cause 63% of deaths globally (23); while in Canada,
these same four diseases account for approximately 75% of all deaths (24). Overall, the
economic burden of chronic disease in Canada is estimated to exceed $93 billion per year (24).
The burden of chronic non-communicable diseases in Canada, and internationally, has led to
calls to address the common risk factors that contribute to chronic disease (2,3,25).
A common modifiable risk factor shared by many chronic diseases, including obesity, diabetes,
cardiovascular diseases, cancer, dental diseases, and osteoporosis, is an unhealthy diet (1). In
order to prevent these diet-related chronic diseases, the World Health Organization recommends
that healthy populations limit their intake of saturated fat, trans fat, cholesterol, sugar, sodium,
and total energy, while ensuring adequate intakes of fat, polyunsaturated fat, carbohydrate,
protein, and fibre (1,3). Similarly, 2007’s Eating Well with Canada’s Food Guide recommends
Canadians limit foods and beverages high in calories, fat, sugar or salt, as well as trans fat while
meeting their needs for vitamins, minerals, and other nutrients (4).
It is widely accepted that a reduction in deaths due to chronic disease can be achieved, in part,
through investments in population interventions to promote healthy diets, such as the eating
patterns proposed by the World Health Organization and Eating Well with Canada’s Food Guide
(23). Among the main interventions proposed to promote healthy diets are food labelling and
restrictions on the marketing of unhealthy food products (25).
4
2.2 An overview of nutrition information on food labels
2.2.1 Nutrition labelling
Food labels are a means through which product information is communicated to consumers (26).
According to the Canadian Food Inspection Agency, a food label serves three primary functions:
1. It provides basic product information;
2. It provides health, safety, and nutrition information; and,
3. It acts as a vehicle for food marketing, promotion and advertising.
To date, food labelling interventions that promote healthy diets to reduce the risk of chronic
disease have focused on the food label’s second function. According to the World Health
Organization, “consumers require accurate, standardized and comprehensible information on the
content of food items in order to make healthy choices” (3). To provide consumers with this
nutrition information, the World Health Organization suggests governments require that
information on key nutritional aspects be provided on food labels – what is commonly known as
nutrition labelling. The Codex Alimentarius Commission recommends that mandatory nutrition
labelling be presented in a tabular format that displays information on nutrients including:
energy, protein, available carbohydrate, fat and saturated fat, sodium, and total sugars (27).
In keeping with the recommendation of the World Health Organization, mandatory nutrition
labels are now imminent or in place in over 20 countries world wide and voluntary nutrition
labelling standards are in place in at least 10 more (15). Within Canada, mandatory nutrition
labelling standards were passed in 2003, 15 years after voluntary nutrition labelling guidelines
were first introduced (5). Mandatory nutrition labelling came into full effect in 2007 and
required the declaration of energy and 13 core nutrients per serving in a Nutrition Facts table
(NFt) on almost all prepackaged foods with a few exceptions.
2.2.1.1 Nutrition labelling: Use and understanding
Since mandatory nutrition labelling came into full effect in 2007, the number of Canadians who
report often or always using the NFt to inform first time purchases has been on the rise (28-30).
The most recent data show that 74% of Canadians are often or always using the NFt for first-time
purchases (30). This finding is consistent with self-reported data from other countries with
nutrition labels, that also reported high levels of use (31-33). However, effective use of the NFt
5
in food choice requires understanding, and research on consumers’ abilities to understand
nutrition labelling is mixed. A few studies have found that consumers, and consumers with low
levels of health literacy in particular, perform poorly when their food label knowledge is tested
(34-36); others have found that, when tested, a large proportion of consumers can successfully
use the nutrition label to find needed information (29,30,35,37-39).
The quantitative portions of nutrition labels, including the percent daily value and amount of
food, appear to be a barrier to many consumers’ understanding and interpretation
(11,29,30,34,35,38-53). Numerous studies have found the percent daily value – the proportion of
daily requirements for a given nutrient provided in a serving of the food – to be a difficult
concept for consumers (11,29,30,34,39-42,45,48-51,54). Indeed, in 2012 only 60% of Canadians
reported feeling confident in their ability to use the percent daily value when choosing foods –
this figure fell to 46% among consumers with less than a high school education (30).
Furthermore, fewer than half of all consumers can determine what is ‘a little’ or ‘a lot’ of a given
nutrient using the percent daily value (29,30). With respect to using nutrition label information
on the amount of food, numerous studies have found consumers struggle to compare two foods
with different serving sizes (29,30,39,46,48), as well as to determine the amount of different
nutrients found in servings larger or smaller than the amount of food depicted in the nutrition
label (35,38,39,44,51-54). For example, one study of Canadian parents found that only 16.4% of
parents could estimate the number of calories in an entire bottle of Cola using a NFt that gave the
nutrition information per serving (52). Consumers with the lowest levels of education appear to
struggle most with nutrition label numeracy (29,30,38,55). In addition to the numeracy
challenges, consumers have also reported difficulty with nutrition label terminology, which is
perceived by some as too technical and advanced (10).
Difficulties understanding and interpreting nutrition labels, like those noted above, have led
expert groups, such as Canada’s House of Commons Standing Committee on Health, to call for a
simplified nutrition labelling scheme (13,56-58). Indeed, research has shown that simplified
labels promote more accurate evaluations of foods by consumers (55,59,60); and, several studies
have reported a benefit to simplified nutrition labels that employ graphics and symbols
(10,33,60-65). Further, research has suggested that consumers may be better served by nutrition
information on the front of food packages rather than on the back, as is the current practice
(10,63,66-68).
6
2.2.2 Nutrition marketing
As noted in the previous section, food labels can function as a vehicle for food marketing,
promotion, and advertising (26). Nutrition marketing has been defined as “any marketing … of
food and beverages using health or nutrition information beyond minimum requirements”
(69,70). In the case of food labels, the NFt constitutes the minimum requirements and any
additional health and nutrition information provided beyond the NFt would therefore be
considered nutrition marketing. Within Canada, voluntary health and nutrition information on
food packages falls under two umbrellas, nutrient content claims and health claims (Figure 2-1).
Figure 2- 1. Nutrition information on prepackaged foods in Ca nada
Nutrition information on Canadian food
labels
Mandatory
List of ingredientsNutrition Facts
tableNutrient content
claims
Disease risk reduction and
therapeutic claimsFunction claims
Nutrient function claims
Strain-caims
1. Nutrition information on prepackaged foods in Ca nada
Voluntary
Nutrient content claims
Health claims
Function claims
Probiotic claims
-specific caims
Non-strain-specific claims
General health claims
Body weight claims
Third-party endorsements and
logosHeart symbols
7
Heart symbolsGuidance for
healthy eating
8
2.2.2.1 Nutrient content claims
Nutrient content claims describe the amount of a nutrient in a food and fall under Canada’s Food
and Drug Regulations (5). The Regulations define a closed list of nutrient content claims with
specific compositional criteria in place for their use. For example, to carry the claim “free of
sodium or salt” a food must contain less than 5 mg of sodium per reference amount and serving
of a stated size, or less than 5 mg of sodium per serving of a stated size if the food is a
prepackaged meal. Additional examples of nutrient content claims include “low in fat”, “a
source of calcium”, and “zero trans fat”, and are listed in a Table following section B.01.513 of
the Food and Drug Regulations.
2.2.2.2 Health claims
Health claims describe (or imply) a relationship between eating a food, or a food ingredient, and
health, and are subject to varying degrees of regulation in Canada (26). Health claims have been
classed into three main categories (Figure 2-1). The first category is disease risk reduction and
therapeutic claims, which fall under Canada’s Food and Drug Regulations. Disease risk
reduction claims describe a link between a reduced risk of developing a diet-related disease, and
the consumption of a given food, or food constituent, as part of a healthy overall dietary pattern.
Presently, five disease risk reduction claims – with prescribed wording – are permitted under the
Regulations (71). For example, “A healthy diet with adequate calcium and vitamin D, and
regular physical activity, help to achieve strong bones and may reduce the risk of osteoporosis”
(26). Therapeutic claims are about treating or easing disease or health conditions, or about
repairing or altering body functions. Presently, six therapeutic claims are permitted (71). For
example, “16 g (2 tablespoons) of ground flaxseed supplies 40% of the daily amount shown to
help lower cholesterol” (72). Prior to permitting the use of a disease-risk reduction or
therapeutic claim on food labels and advertisements, Health Canada completes a review of the
scientific evidence for the claim (71). Only claims substantiated by the evidence are permitted.
Further, to use a given disease risk reduction claim or therapeutic claim on its label, a food must
meet prescribed conditions related to energy and/or other nutrients relevant to the claim (26).
The second category of health claims is function claims (26). Function claims relate the positive
contribution of eating a food, or food constituent, to normal body functions and biological
activities (e.g. health, maintaining physiological functioning, physical or mental performance).
9
These claims cannot refer to the treatment or prevention of any disease or of the symptoms of a
disease. While function claims must comply with subsection 5.1 of Canada’s Food and Drugs
act, that forbids “false, misleading, or deceptive” product labelling (73), there are presently no
specific regulations governing their use (26). However, the Canadian Food Inspection Agency’s
Guide to Food Labelling and Advertising has set guidelines for their use. “Consumption of 1
cup (250 ml) of green tea increases antioxidant capacity in the blood” is an example of a function
claim. Nutrient function and probiotic claims are subsets of function claims. Nutrient function
claims describe the generally accepted role of energy or nutrients in maintaining health, or in
normal growth and development. Unlike more general function claims, nutrient function claims
are provided for under the Food and Drug Regulations; the Regulations set out minimum
requirements for the food’s content of protein, or specific vitamins or minerals relevant to the
claim. “DHA, an omega-3 fatty acid, supports the normal physical development of the brain,
eyes and nerves primarily in children under two years of age” is an example of a nutrient
function claim. Probiotic claims are statements about the health benefits or effects of specific
probiotic strains (strain-specific claims) or the nature of probiotics (non-strain specific claims).
Any probiotic claim that is therapeutic in nature would be subject to the Food and Drug
Regulations, however there are presently no such claims approved for use in Canada. Aside
from that, there are no specific regulations for the use of probiotic claims, however guidelines
have been described in Health Canada’s The Use of Probiotic Microorganisms in Food (74).
These guidelines include a requirement that the food contain at least the minimum amount of
probiotic required to achieve the claimed effect. “Provides live microorganisms that naturally
form part of the gut flora” is an example of a probiotic claim (26).
The final category of health claims is general health claims (26). General health claims are
broadly defined as claims that “promote health through healthy eating or that provide dietary
guidance”. Such claims do not discuss health effects or conditions, or disease and are generally
not subject to any specific regulations. However, they are subject to subsection 5.1 of the Food
and Drugs Act and as such may not be “false, misleading, or deceptive” (73). Further, guidance
for their use is given in the Canadian Food Inspection Agency’s Guide to Food Labelling and
Advertising (26). The Guide advises on general health claims that relate to body weight, third-
party endorsements and logos, heart symbols, and guidance for healthy eating. For instance,
general health claims may state that a food is intended for use in maintaining body weight (e.g.
10
“As part of healthy eating, this food may assist in achieving and maintaining a healthy body
weight because it is lower in energy”), provided the food’s label meets prescribed conditions.
Health professionals or organizations, individuals and groups may give third-party endorsements
or logos to products, so long as they do not mislead or confuse consumers as to the advantages of
the food. According to the Guide, to avoid such confusion a third-party endorsement or logo
should adhere to the following principles:
1. Does not give the impression that a single food or brand of food is "healthier" than, or
nutritionally superior to, other foods not bearing the third party's name, statement, logo,
symbol, seal of approval or other proprietary mark. …
2. Does not give the impression that the food is a treatment, preventative or cure for
disease. A third party's name, statement, logo, etc. must not suggest that a food may
prevent, cure or treat a disease … Such a suggestion is false and specifically prohibited
by the Food and Drugs Act (26).
General health claims in the form of heart symbols are generally not acceptable, but they can be
used in conjunction with third-party endorsements and logos or with disease-risk reduction
claims related to heart health (26).
2.2.2.3 Nutrition marketing: Use and Understanding
Surveys of packaged foods sold in North American have found that nutrition marketing
messages are common on food labels (70,75-78). Recent estimates place nutrition marketing
information on 48.1% of Canadian (78), 44.3-71.0% of American (70,75), and 14% of Australian
packaged food labels (79). In addition to widespread use by manufacturers, a small but
significant proportion of consumers appear to be seeking out nutrition marketing information on
food labels. In 2008, 21% of Canadian label readers looked for nutrient content claims and 16%
looked for health claims (6). European estimates are even higher, with 59% of consumers
reportedly ‘often’ or ‘always’ reading claims (80). Moreover, a recent ethnographic study found
that Canadian consumers made use of front-of-package nutrition marketing information, like
nutrient content claims, more often than the NFt (81).
Widespread use of nutrition marketing information is of interest because nutrition marketing
could potentially influence food purchases, which may impact consumption patterns and finally
chronic disease risk (69,70). The presence of nutrition marketing information on food labels has
11
been shown to influence consumers’ perceptions of food products’ healthiness and nutritional
quality (80,82-89). In fact, the influence of nutrition and health claims on consumers’
perceptions of healthiness and nutritional quality have been found to extend beyond the
nutritional and health elements identified in the claim – what is known as the ‘halo’ effect
(82,83,86,88,89). For example, consumers rated the same bread lower in calories when they saw
a nutrition marketing claim stating that the product was low in carbohydrates than when they saw
no such claim (86). In addition, consumers perceived a cereal as nutritionally superior, as well
as more beneficial to digestive health in the presence of a fibre claim compared to the same
cereal without a fibre claim; they also believed that their likelihood of disease risk and weight
gain was lower (88). Similarly, consumers rated a soup as overall healthier, and more beneficial
for hypertension, high cholesterol, heart problems, constipation and diabetes when either a
nutrient-content, function, or disease-risk reduction claim related to sodium was present; despite
the fact that all soups had the same NFt and the claims only mentioned a relationship between
sodium and blood pressure (89). This is problematic because, as previously noted, in order to
carry nutrient-content, disease-risk reduction, therapeutic, nutrient function, and probiotic claims,
foods need only meet compositional criteria related, or relevant to, the nutrient or food
component that is the subject of the claim (26,74,90). Further, with respect to disease risk
reduction and therapeutic claims, the relationship between the nutrient or food components that
are the subject of the claim and health have only been substantiated for the disease or condition
cited in the claim (71). Therefore it would appear that these nutrition marketing claims might be
giving consumers erroneous impressions about the nutritional and health qualities of the products
that carry them. This is problematic given that the presence of nutrition marketing claims, and
health claims in particular, have led consumers to report greater intentions to purchase a product
(80,83,84,89,91); however, it is unclear if greater purchase intentions are translated into actual
in-store purchases.
2.3 An overview of FOP nutrition rating systems and symbols
Yet another form on nutrition marketing found on food labels, not yet discussed, is front-of-pack
(FOP) nutrition rating systems and symbols. FOP systems and symbols summarize key
nutritional aspects and characteristics of food products and are typically found on a product’s
principal display panel, but may be elsewhere on the food label (19). FOP systems and symbols
12
often integrate characteristics of traditional nutrition labels, nutrient-content claims, and even
disease-risk reduction, therapeutic, and function claims.
2.3.1 History of FOP symbols
FOP nutrition rating systems and symbols date back to the 1980s, when the American Heart
Association initiated its Heart Check symbol (1987) and the Swedish National Food
Administration introduced its Nordic Keyhole (1989) (19). Canada’s first FOP symbol, the
Heart and Stroke Foundation of Canada’s Health Check™, was introduced in 1999 to help
consumers identify healthier food choices. Typically, manufacturers receive the right to use
third-party symbols by paying a program fee (92). Although non-profit health organizations and
governments developed the earliest entrants into the FOP symbol marketplace, industry quickly
followed suit with food manufacturers, retailers, industry and non-industry associations, and non-
industry experts all introducing their own symbols in the years that followed (13,19,22,92,93).
For instance, in 2005 PepsiCo expanded its Smart Spot symbol into Canada and was quickly
followed by a number of other proprietary FOP systems developed by food manufacturers (e.g.
Kraft Sensible Solutions, President’s Choice Blue Menu, Kellogg’s Get-the-Facts, etc.) to help
consumers identify their “healthier” products.
In the only systematic study of the use of FOP symbols on food packages completed to date, our
research group at the University of Toronto identified 158 unique FOP symbols in use on
packaged food labels in Canada in 2010-2011 (78). Overall, one or more FOP symbols appeared
on 18.9% of food products. The number of FOP symbols in use internationally has not been
comprehensively studied. However, reports have described 8-30 different FOP nutrition rating
systems or symbols in use internationally (19,92-96). Further, the overall prevalence of FOP
systems on food products in markets outside of Canada has not been studied. Point-of-purchase
research emanating from the Netherlands, United Kingdom, and Australia has found that some
FOP systems can be found on up to 18% of products purchased by consumers (97,98). However,
the proportion of products carrying a FOP system in these studies may not reflect the prevalence
of such products in the marketplace, but rather reflect the popularity of products bearing FOP
systems.
13
2.3.2 FOP nutrition rating system and symbol categories
In 2010, the United States (US) Institute of Medicine classified the myriad FOP systems and
symbols in use internationally into three categories (19):
1. Nutrient-specific systems;
2. Summary indicator systems; and,
3. Food group information systems.
The sections that follow describe these three categories and examples of each FOP type are
found in Figure 2-2.
Figure 2-2: Examples of different front- of
Nutrient Specific Systems
Nutrient- specific systems that display the amount of calorie s and select nutrients per serving
The Grocery Manufacturers Association and the Food Marketing Institute’s Facts Up Front
Nutrient- specific systems based on claim criteria
General Mills’ Goodness Corner
of-pack symbol types
Nutrient Specific Systems
specific systems that display the amount of calorie s
The Grocery Manufacturers Association and the Food
specific systems based on claim criteria
Summary Indicator Systems
Heart and Stroke Foundation of Canada’s Health Check™
Kraft’s Sensible Solutions™
14
Heart and Stroke Foundation of
Food Group Information
Systems
Whole Grain Council’s Whole Grain Stamp
15
2.3.2.1 Nutrient-specific systems Nutrient-specific systems include 1) systems that display the amount per serving of calories or
select nutrients from the NFt on the FOP, and 2) symbols based on nutrient-content or health
claim criteria (19) and appear on 4.9% of packaged Canadian food products (78). Nutrient-
specific systems are based in the regulations for nutrition labelling and claims (19). Systems that
display the amount per serving of calories and/or select nutrient on the FOP are closely related to
mandatory back-of-pack nutrition labels, such as the NFt. European researchers sometimes refer
to these systems as “banding” or “nutrient signposting” (93). These systems are designed to be
comprehensive and appear on as many foods as possible. In addition to displaying the amount of
calories and select nutrients per serving, these systems may also display percent daily value
information (19). Systems displaying percent daily value information are known as percent-
daily value systems, guideline daily amount systems, or daily intake guide systems depending on
the country, however for simplicity they will be referred to hereafter as daily value systems.
Daily value systems have been found on up to 67% of packaged food labels in some countries
outside of Canada (99,100). Systems that display the amount per serving of calories or select
nutrients may also include colours (traffic lights) or words that interpret for the consumer if the
specific nutrient amounts found in the product are “high”, “medium”, or “low” (19). In 2008-
2009, traffic light systems were found on ~3% of packaged food labels in the United Kingdom –
the country where they originated (100). An example of a system that displays the amount of
calories and select nutrients per serving is the US Grocery Manufacturers Association and Food
Marketing Institute’s Facts Up Front (101). Facts Up Front displays on a food label’s principal
display panel the amount of calories, fat, sodium, and sugar, plus two additional nutrients, per
serving of a food. When applicable, Facts Up Front also provides percent daily value
information.
Symbols based on claims would be considered nutrient-content or health claims (depending on
the claim they portray) in addition to being FOP symbols. As a result, they would be subject to
the same regulations described in section 1.2.2.1 or 1.2.2.2, requiring them to meet the thresholds
established per reference amount and serving of a stated size for the nutrient-content or health
claim depicted in the symbol. Multiple symbols based on claim criteria may be placed on a
single product depending on which nutrients are included in the system (19). An example of
such a system used in Canada is General Mill’s Goodness Corner which awards foods multiple
16
symbols based on their compliance with nutrient content claims such as “low fat”, “excellent
source of iron”, and “high fibre” (94).
2.3.2.2 Summary indicator systems
Summary indicator systems, also known as signpost or health logos, give no specific nutrient
content information to consumers, but instead use a single symbol, icon, or score to provide
summary information about the nutrient content of a food (19). In Europe, summary indicator
systems are known as “point of purchase” systems or “integrative approaches” (93). FOP
systems within this category are typically only applied to ‘healthy’ or ‘healthier’ choices within a
product category. This type of symbol has been identified on 7.5% of packaged foods sold in
Canada (78), and fewer 2% of products in Europe (100). Summary indicator systems attempt to
assess the overall healthiness of a food using either thresholds or algorithms (19). Threshold-
based systems typically establish maximum levels for nutrients to limit, and minimum levels for
nutrients or food components to encourage, to judge whether a product qualifies for a summary
indicator symbol (21,22). Algorithm-based systems award points for the presence of nutrients or
food components to encourage and subtract them for the presence of nutrients to limit to arrive at
a final score that is used as the summary indicator symbol (19,21). Summary indicator systems
are considered to be general health claims, and as such are not subject to any specific regulations.
However, as noted in section 1.2.2.2, they may not be “false, misleading, or deceptive” (73) and
they should follow all relevant sections of The Guide to Food Labelling and Advertising, such as
the guidance on the use of third-party endorsements and logos (26). An example of a summary
indicator system based on a threshold model is the Heart and Stroke Foundation of Canada’s
Health Check™ (102). The Health Check™ system awards a single symbol to the products of
participating manufacturers that meet Health Check™’s category-specific threshold criteria
established by Registered Dietitians based on Eating Well with Canada’s Food Guide and
nutrient-content claim criteria. An example of a summary indicator system based on an
algorithm model is NuVal ®(103). Using an algorithm that takes over 30 nutrients and food
components into account, NuVal ® gives food a single score from 1-100. One limitation of
summary indicator systems is that consumers cannot tell at the point-of-purchase the role
different nutrients and food components played in the food’s evaluation, and for some systems
(e.g. NuVal ®) full details of the scoring system are not publically available (19).
17
2.3.2.3 Food group information systems
Food group information systems include symbols that indicate that a food group (such a
vegetables and fruit) or a food ingredient important to the diet (such as whole grains) is present
in a food product (19). Some of the food group symbols reference the presence of a serving (or
partial serving) of a food group from Eating Well with Canada’s Food Guide (using Food Guide
standards of what constitutes a serving). This type of symbol appears on 3.5% of packaged
foods sold in Canada (78). An example of a food group information symbol is the Whole Grain
Council’s Whole Grain stamp (104).
In addition to the three categories of FOP symbols described by the Institute of Medicine, our
research group identified yet another category in our study on the use of FOP symbols on 2010-
2011 food packages: hybrid systems (78). Hybrid systems combined features of two or more of
the categories previously described and appeared on 7.0% of all packaged foods products in
Canada.
2.3.3 Nutrient profiling
Underpinning each FOP nutrition rating system, regardless of category, is a nutrient-profiling
scheme (19,93). Nutrient profiling is the classification of foods based on their nutritional
composition (95,105), however in the absence of specific regulations for their use, FOP symbols
in Canada do not adhere to any standardized nutrient profiling criteria (13). While most FOP
symbols share an overarching intent to help consumers quickly compare foods, identify
nutritious choices, and determine if a food meets their nutrient needs, they each have their own
purpose and target-audience (19). In addition, the nutrient profiles underlying different FOP
systems may vary in their: 1) application of nutritional criteria; 2) selection of nutrients and other
food components included; 3) amount of food; 4) type of model; and, 5) criteria’s basis
(19,20,93-95,105,106). Indeed, the US Institute of Medicine has stated that no two FOP symbols
share the same underlying nutrient profile (19).
Application of the criteria refers to whether or not the system applies criteria universally (across-
the-board) or by product category (19,22,93,95,105,106). Category specific criteria recognize
naturally occurring differences in nutrient profiles by product category and allows consumers to
identify healthier selections within a product category (19,22,107). However, a category-based
system requires the development of food categories, which is not always evident (19). Universal
18
criteria apply the same criteria to all foods, and allow consumers to compare foods across all
categories. Universal criteria also allow FOP systems to use existing government regulations for
nutritional labelling and nutrient content claims as the basis for the criteria (19). Summary
indicator systems tend to use category specific criteria over universal criteria; whereas nutrient-
specific and food group information systems tend to apply their criteria universally. Regardless
of whether a FOP system’s nutrient profile is category specific or universal, one study has found
that consumers believe that all FOP systems compare products across all product categories.
Feunekes et al (63) reported that 58% of European consumers thought that FOP systems
compared products across the board, while only 27% thought they compared products within
categories.
There is tremendous variety in the nutrients and food components included in the nutrient
profiling schemes that are used from one FOP system to the next. Reviews have identified FOP
systems based on as few as one or two nutrients and/or food components to more than 30
(19,92,93,103,106,108). Nutrients incorporated in one or more FOP system included energy,
total fat, saturated fat, trans fat, cholesterol, omega-3 fats, carbohydrate, starch, added sugar,
fibre, whole grains, protein, sodium, and vitamins and minerals (vitamin A and C, folate, iron,
calcium). Most nutrient profiles include a combination of nutrients to limit (“negative”
nutrients) and nutrients and/or food components to encourage (“positive nutrients”), although
some systems focus on just one or the other (19,92,105,106). Nutrients to limit have been
defined variously as calories, fats and oils (total fat, saturated fat, trans fat, cholesterol), added
sugars, and sodium (19,22,105). Nutrients and food components to encourage have been defined
variously as vitamins and minerals (such as calcium, potassium, magnesium, and vitamins A, C,
and E), fibre, fruits and vegetables, whole grains, and low-fat dairy. Depending on the system,
the nutrients included in the profile may be based on existing dietary guidance from government
or other authoritative bodies, or a scientific advisory panel put together by the system proprietor
(19). The nutrients and/or food components included in a FOP system should be those that have
been acknowledged as important in consensus documents developed by authoritative bodies,
however, many systems include nutrients and food components for which there is insufficient
evidence to support their inclusion. In 2011, the US Institute of Medicine concluded that
saturated and trans fat, sodium, and added sugars are the nutrients of greatest public health
concern, and therefore should be included in the nutrient profile of FOP systems (19,109).
19
Nutrition labelling in Canada is based on reference amounts and servings of a stated size (26). A
reference amount is a regulated amount of a type of food typically eaten by an individual in one
sitting and is the basis for nutrient-content and health claims in Canada. A serving of a stated
size is the amount of food forming the basis for the nutrition information found in the NFt.
Manufacturers are allowed to select from a range of serving sizes – which encompass the
reference amount – for use in the NFt depending on the product. FOP systems may be based on
reference amounts, serving size, or other amounts of food such as 100 g, 100 ml, or 100 calories,
or some combination of these (19,20,95,106). Stockley’s (93) review of FOP symbols in
Europe, Australia, and North America found that most schemes developed by public sector
interests (like health organizations, governments, and nutrition councils) used a standardized
amount, like per 100 g, as their base, whereas commercial interests (like manufacturers and
retailers) appeared to use both per serving and per standardized amounts equally.
Two types of nutrient profiling models are used in FOP systems: thresholds and algorithms
(108). As previously described, threshold criteria typically establish maximum levels for
nutrients to limit, and minimum levels for nutrients or food components to encourage (21,22),
whereas algorithm models award points for the presence of nutrients or food components to
encourage and subtract them for the presence of nutrients to limit to arrive at a final score
(19,21). Nutrient-specific systems based on claims criteria, or that add interpretive elements
(such as colours or words) to information on the amount of calories or select nutrients per
serving, use thresholds, as do food group information systems (19). Summary indicator systems
may use either thresholds or algorithms. Reviews of FOP nutrition rating systems have revealed
that threshold criteria are the dominant model for nutrient profiling (19,93,105).
Finally, the basis for the numbers used in nutrient profiling criteria differs from one FOP system
to the next. A number of FOP systems base their criteria on government nutrition labelling
regulations, dietary guidance from government or other authoritative bodies, or on the opinions
of scientific advisory panels (19,93). However, for many FOP systems used in Canada and
abroad the basis of their nutrient profiling criteria is not apparent (92,93,105). This is especially
true for commercial FOP systems.
Presently there is no single approach to nutrient profiling that is generally accepted for
classifying foods for FOP systems or other purposes. Moreover, the World Health Organization
20
and other experts have raised concerns that multiple different nutrient profiling models, as are
seen in existing FOP systems, can lead to consumer confusion and mistrust (13,56,92,94,95).
Further, the nutrient profiling criteria underlying many popular FOP systems have been criticized
as being too relaxed (i.e. highly processed foods high in sugar and low in whole grains may
qualify for symbols) (56). For example, the industry developed Smart Choices summary
indicator system came under scrutiny when it was shown that products such as Froot Loops and
Cocoa Krispies qualified for this symbol on the basis of the system’s nutrient-profiling criteria
(56,110). Further, the nutrient profiling criteria of the majority of FOP systems is undocumented
in the literature, and fewer still have been empirically evaluated or validated (92,105), making it
difficult to determine if these systems are truly directing consumers to healthier choices. As a
result, there have been calls from a public interest group for the development of minimum
standards to ensure consistency across voluntary FOP symbols (109). In 2014, Health Canada
released a nutrient profiling system to classify foods from the four food groups into four tiers:
“Foods in line with Canada’s Food Guide guidance” (tiers 1 and 2); “Foods partially in line with
Canada’s Food Guide guidance” (tier 3); and, “Foods not in line with Canada’s Food Guide
guidance” (tier 4) (111). While this system has the potential to be applied as a minimum
standard for FOP symbols, at present Health Canada describes these criteria as exclusively a
surveillance tool.
2.4 FOP systems and the nutritional quality of foods
2.4.1 FOP systems and reformulation
FOP systems have been proposed as a potential public health intervention to improve population
level dietary intakes (19); one way in which FOP systems are anticipated to do so is by
stimulating manufacturers to reformulate their products in a more healthful way in order to meet
the FOP system’s nutrient profiling criteria. The current evidence, although limited in both
quantity and quality, suggests that the nutrient-profiles underlying selected summary indicator
systems are in fact stringent enough to stimulate healthier product reformulations by
manufacturers (112-115). Researchers in Australia and New Zealand, Canada, and the
Netherlands have found that summary indicator systems have been successful in stimulating
participating manufacturers to lower their foods’ sodium, saturated fat, trans fat, and/or calorie
21
content and/or increase their fibre content, across a range of product categories, or formulate new
products in order to meet the system’s nutrient profile. For example, in order to qualify for the
Choices logo, participating companies in the Netherlands were able to reformulate processed
meat products and reduce their saturated fat content from 3.09 g to 1.74 g and their sodium
content from 1018 mg to 835 mg per serving (114). Similarly, Canadian manufacturers
participating in the Health Check™ program were able to achieve a reduction of 80-150 mg of
sodium per reference amount in reformulated products in categories ranging from dinners and
entrees to deli meats (115). Although these findings are promising, because the FOP systems in
question are voluntary, and sometimes require the manufacturer to pay a licensing fee for their
use, it is unclear whether competitors not participating in these programs undertook similar
reformulations over the same period of time. Furthermore, the products of manufacturer’s not
participating in these programs have not been evaluated against the systems’ nutrient profiling
criteria. As a result, it is unclear how many of said products would qualify for these FOP
systems as is, versus require reformulation should they choose to participate in the program.
Thus the potential for summary indicator systems to stimulate reformulation by prospective
program participants is unclear.
Only one study to date has examined the potential impact of nutrient-specific systems on
reformulation. A recent study analyzed the nutritional quality of breakfast cereals over a period
of six years (2004-2010), during which time a voluntary daily value system – Australia’s Daily
Intake Guide – was introduced (2006) (116). The study compared the nutrient content of cereals
year over year as well as the nutrient content of cereals with and without the Daily Intake Guide
FOP system. Of the 67 cereals available as the same product in 2004 and 2010, 60 had changed
their formulation; however, this did not result in significant changes in their nutrient content,
aside from a small increase in the protein content of cereals without the FOP system, leading the
authors to conclude that nutrient-specific systems, like the Daily Intake Guide, that provide
information on calorie and nutrient content per serving on the FOP without interpretive
information do not appear to promote product reformulation. Similar data on reformulation is
unavailable for other nutrient-specific systems (those based on claims or with interpretive
elements) and food group information systems.
22
2.4.2 The nutrient profiles of foods with and without FOP systems
Experts in Canada have raised concern that consumers may, by default, perceive foods without
FOP symbols as less healthy than foods with symbols (78,81), despite the fact that there no
evidence to suggest that foods with symbols have lower levels of nutrients of public health
concern. In the current Canadian environment where multiple proprietary and third-party FOP
symbols are in use, not all products on the grocery shelves have been evaluated using the
available FOP systems. Therefore, many products may qualify for one or more FOP symbols but
may not actually use these symbols on their labels, making it difficult for consumers to use these
symbols to compare products (19). For example, all products carrying the Heart and Stroke
Foundation of Canada’s Health Check™ symbol meet the Health Check™ criteria, but not all
products that meet the Health Check™ criteria carry the symbol. As a result, consumers cannot
be sure if a product does not carry the Health Check™ symbol because it did not meet the
system’s nutrient profiling criteria or simply because its manufacturer is not participating in the
program. Moreover, manufacturer developed FOP symbols may only be carried by qualifying
products of the manufacturer who developed the system in question, and as a result consumers
cannot be certain products without a symbol are in fact of a lower nutritional standard or if that
manufacturer simply does not have its own proprietary symbol (92). However, just how many
products in the Canadian marketplace have been excluded from carrying specific FOP systems
for reasons unrelated to their nutritional composition has not been evaluated. As a result, it is
unclear if Canadians can rely on specific licensed or proprietary FOP symbols to identify
products of superior nutritional quality.
As previously noted, the nutrient profiling criteria behind the FOP symbols found on food labels
differ from one system to the next. However, the details of these nutrient profiles are invisible to
the consumer at the point of purchase (117). In environments where a single FOP system has
been universally applied (i.e. all products sold in store are evaluated against the same nutrient
profiling criteria and symbols are awarded to all qualifying products), one study has shown that
consumers can rely on FOP symbols as an indicator of products with lower levels of negative
nutrients and higher levels of positive nutrients. Sutherland et al. (118) found that breakfast
cereals bearing the Guiding Stars summary indicator system, which was universally applied to
all products sold at a Northeastern US grocery chain, had less sugar and sodium, and more fibre
than cereals without Guiding Stars. However, more research is needed to determine if their
23
conclusion can be applied to other universally applied FOP systems. Moreover, the current
Canadian grocery marketplace is not characterized by the use of a single FOP system,
universally applied to all products, but rather the use of multiple FOP systems with distinct
nutrient profiling criteria. Lack of specific regulations governing the use of these FOP systems
means that there is no minimum nutrient profile that foods must meet to carry a FOP symbol.
Without minimum nutrient profiling standards, products that qualify for one FOP symbol may
not qualify for a different FOP symbol. For instance, products that have failed to meet the
nutrient profiling criteria for a third-party FOP symbol, like the Heart and Stroke Foundation’s
Health Check™, may qualify for a symbol developed by their own manufacturer whose nutrient
profiling criteria is less stringent (92). Indeed, different FOP systems do not show perfect
agreement in which foods can qualify for a symbol. A European study that applied the nutrient
profiling criteria of three FOP systems to the same foods found only moderate agreement (41-
60% identical ratings) between how some of the FOP systems classified the products (108). As a
result, it is unclear if consumers can rely on the mere presence of any FOP symbol as an
indicator of products that meet a higher nutritional standard.
In 2011, the US Institute of Medicine stated that in order to best promote the health of the public,
FOP systems must take into account nutrients and nutritional components of greatest relevance
to chronic disease risk (109). This led them to conclude that calories, saturated and trans fat,
sodium, and added sugar should be considered in the nutrient profiles of FOP systems. Despite
this, most countries – Canada included – allow products to carry nutrient-specific symbols
without considering their overall composition with respect to calories and these four nutrients of
public health concern (119). Indeed, nutrient-specific systems based on claim criteria have
caused concern that some products qualifying for a nutrient-content or health claim for one
nutrient may not be “low” in other nutrients that should be limited (19,108). Similarly, food
group information systems typically only consider a single food group or ingredient and not
overall nutrient content (19). Further, although summary indicator systems consider multiple
nutrients and food components in their nutrient profiling criteria, the elements considered differ
from one system to the next (19,92,93,106,108), and do not necessarily include calories and the
nutrients flagged by the Institute of Medicine as being important to public health. As a result, it
is presently unclear if FOP symbols in general can be relied upon as a guide to foods with more
24
favourable levels of calories and nutrients related to chronic disease risk (i.e. saturated fat, trans
fat, sodium, and added sugar).
2.5 Influence of FOP systems on perceptions of nutritional quality
Despite the fact that the overall healthiness and nutrient content of foods with FOP symbols is
unclear, evidence suggests that consumers perceive products with certain categories of FOP
symbols as being overall healthier, as well as having superior nutrient profiles (120-124). With
respect to nutrient-specific systems, one study has demonstrated that even the presence of a FOP
system that simply provides information on the content of calories and select nutrients per
serving increases consumers’ product healthfulness perceptions and purchase intentions over
conditions where the same product is shown without this nutrient-specific system (124).
Moreover, research has show that when an element of interpretation, such as traffic light colours,
is added to a nutrient-specific system that displays the amount of calories and selected nutrient
per serving, consumers’ perceptions of healthfulness and nutrient content are influenced
favourably (123). In a 2011 study, consumers were randomized to view a frozen meal with or
without a traffic light nutrient-specific system that rated the food’s sodium and cholesterol
content as high (red), calorie and sugar content as medium (amber), and fat and saturated fat
content as low (green) (123). Consumers exposed to the traffic light rated the frozen meal as
significantly lower in calories and the nutrients given green or yellow traffic lights. These
favourable evaluations even extended to nutrients not included in the traffic light symbol –
consumers exposed to the traffic light also rated the meal as being lower in trans fat and
carbohydrates. Ratings of the meal’s content of nutrients given red traffic lights did not differ
between the two conditions, however consumers who saw the traffic light system expressed
greater intentions to purchase the product.
With respect to nutrient-specific symbols based on claims criteria, given that these symbols are
forms of nutrient-content and health claims, it follows that these symbols most likely exert the
same effect on consumers’ perceptions of the healthiness and nutrient content previously
demonstrated for claims made on food labels. As previously described in section 1.2.2.3,
consumers perceive products as healthier and of superior nutritional quality when they carry
nutrition marketing messages such as nutrient-content and health claims (80,82-89), and that
25
these perceptions extend beyond just the nutrient that is the subject of the claim (i.e. the ‘halo’
effect) (82,83,86,88,89). Indeed, experts have raised concerns that FOP systems that only
consider a single nutrient in their nutrient profiles (as with some nutrient-specific symbols based
on claims criteria) risk oversimplifying complex nutrition messages (92).
With respect to summary indicator systems, several studies have found that consumers perceive
foods as healthier overall when they carry a summary indicator system compared to when they
do not, even when this perception is not merited by the food’s nutrient content (120-123). For
instance, participants in a crossover experiment who were exposed to the same chocolate mousse
cake on two occasions, once with a summary indicator symbol (Choices) and once without a
symbol, rated the cake as significantly less unhealthy when the symbol was present (122). In
addition, Andrews et al. (123) found that consumer’s randomized to view a frozen meal with a
summary indicator symbol (Smart Choices) rated the meal as significantly healthier than
consumers randomized to see the frozen meal with no symbol. Further, consumers who saw the
symbol in this study also rated the meal as lower in calories, fat, saturated fat, trans fat, sodium,
and sugar than consumers who saw the meal without the symbol. Moreover, consumers who
saw the symbol perceived regular consumption of the meal as less likely to contribute to weight
gain and coronary heart disease risk, and expressed greater intentions to purchase the product.
However, the presence of a mandatory nutrition label alongside the meal with the Smart Choices
symbol increased consumers’ accuracy in evaluating nutrient levels. Another study of the Smart
Choices summary indicator symbol failed to detect any significant differences in consumers’
perceptions of overall healthfulness, and sugar and vitamin content of a breakfast cereal when
the Smart Choices symbol was present versus when it was absent, nor did it lead to self-reported
greater intentions to purchase the product (125). However, it is worth noting that 92% of
participants reported looking at the mandatory Nutrition Facts panel of this cereal (125), and a
moderating effect of the mandatory nutrition label on healthiness evaluations of products with
the Smart Choices symbol was observed in the study by Andrews et al. (123).
Finally, evidence suggests that the effect of summary indicator systems on perceptions of
healthiness and nutrient content is stronger than that of nutrient-specific systems. Andrews et al.
(123) found that while the presence of either a summary indicator symbol (Smart Choices) or a
nutrient specific system (traffic lights) favourably influenced consumers’ perceptions of a frozen
26
meal over the absence of such systems, summary indicator systems influenced favourable
perceptions to a greater extent.
To our knowledge, the impact of food group information systems on consumers’ perceptions of
healthiness and nutrient content has not been studied. Further, at the time this thesis research
was carried out, the impact of different nutrient-specific and summary indicator systems on
Canadians’ perceptions of the healthiness and nutrient content of different foods had not been
evaluated.
While numerous studies have demonstrated that FOP systems influence consumers’ perceptions
of healthiness and nutrient content, and purchase intentions, there is little research actually
demonstrating an effect of FOP symbols on actual purchases and consumption (126). A study of
the Guiding Stars system found that the sales of breakfast cereals with this summary indictor
symbol increased following the system’s introduction in a chain of grocery stores in the
Northeastern US (118). Similarly, the presence of a traffic light system increased the
consumption of foods with “green’ traffic lights and decreased the consumption of foods with
“red” traffic lights in a cafeteria setting (127). In contrast, two studies that examined the effect
of nutrient-specific traffic light systems on product sales found no impact of these symbols on
the sale of ‘healthier’ and ‘less-healthy’ products (128,129). However, it is worth noting that
both these studies were relatively short in duration and only examined the retailer’s own-brand
products from a handful of food categories. Research has found little to no effect of the use of
FOP systems on dietary intakes of individual consumers (64,121). For instance, Borgmeier and
Westenhoefer (64) found five different FOP systems had no impact on energy and nutrient
content of one-day menus selected by German consumers. However, Reid et al (121) found that
self-reported users the Health Check logo consumed 3.5% (p<0.05) fewer calories from fat than
non-users, however it is unclear which behaviour drives the other (130). Similarly, while some
have suggested that the presence of a FOP system on a food label may influence consumers to
consume more of the product than they would if the FOP system was not present, research to
date has not supported this hypothesis (122). While research on the impact of FOP symbols on
product purchases and consumption is limited, Canadian consumers do report looking for these
symbols on food labels. 23% of Canadian label readers report looking for better choice slogans,
symbols or logos [FOP systems] in general on food labels (6), while 23% report occasionally or
regularly looking for the Health Check™ symbol specifically (121). Furthermore, a study out of
27
the Netherlands found that self-reported intentions to purchase products with a summary
indicator symbol, the Choices logo, translated into a larger number of products with the symbol
being purchased at the supermarket (98), suggesting the potential positive influence of FOP
symbols on actual product purchases.
2.6 Is there a need for a single, standardized FOP system?
In section 1.2.1.1 simplified nutrition labels, on the front of food packages, were proposed as a
solution to some of the challenges associated with using the mandatory nutrition label, the NFt,
to choose foods consistent with a healthy diet. FOP systems have been proposed as this
simplified nutrition label (19). However, concerns have arisen that the current FOP system
environment, with multiple co-existing symbols, may be more of a hindrance than help. As
previously noted, experts worry that the presence of multiple FOP systems with different
underlying nutrient profiling criteria may be misleading consumers (13,19,92,94), and there is
also concern that the inconsistent appearance and application of various FOP systems may be
confusing consumers and inhibiting their ability to compare and select foods consistent with a
healthy diet. In fact, following a comprehensive review of the literature on FOP systems, the
Institute of Medicine concluded that multiple systems could confuse the consumer and cause
problems in consumers’ ability to interpret the nutrition information the systems provide (109).
Research conducted in the United Kingdom has found that consumers have difficulties
comparing different types of FOP systems, particularly when the systems shared no common
elements such as gram weights or percent daily values (131). These difficulties were
exacerbated when consumers where asked to compare two products with different FOP systems
where the healthier option was not obvious. Comparing products using different FOP systems
was time consuming for the study participants. While consumers were willing to devote the time
necessary to compare products in this way in the study environment, many indicated that they
would be unlikely to do so in a real-life shopping situation and would look to other information
on the packaging to make their decisions. Qualitative research has found that when comparing
foods with traffic light and daily value systems, consumers sometimes misinterpret the
background colour, such as blue, of monochromatic daily value systems as indicators of low
nutrient levels (132). This is of particular interest as, here in Canada, many manufacturers apply
blue colour schemes in the packaging of the “healthy” lines of their private label brand products
(e.g. President’s Choice Blue Menu or Compliment’s Balance).
28
Concerns about consumer confusion surrounding the use of multiple FOP systems in the same
marketplace has led to expert calls for the adoption of a single, standardized FOP system
(13,56,109). In 2007, the Parliament of Canada’s Standing Committee on Health recommended
that the federal government “implement a mandatory, standardized simple front of package
labelling requirement on pre-packaged foods for easy identification of nutritional value” (13).
Despite this, the Canadian government has taken a hands-off approach to FOP systems, stating
that “it’s great that organizations … have developed systems to help give Canadians even more
information about the food they purchase” and that they do not intend to pursue a single,
standardized FOP nutrition rating system (133).
Beyond the opinions of experts, consumers in many countries have also expressed desire for a
single, standardized FOP system for use on all packaged foods (10,65,131). Confronted with
multiple FOP systems on food products, consumers in the United Kingdom have questioned why
a consistent FOP labelling system has not been adopted by industry (131,132), while, in
Australia, 90% of participants in an intercept survey favoured consistent FOP labelling across all
food products, perceiving it as easier to understand (65). However, to our knowledge,
consumers’ attitudes towards the adoption of a single, standardized FOP system has not been
examined in Canada.
2.6.1 The ideal FOP system for use with consumers
Despite calls for the adoption of a single, standardized FOP system for use in Canada and in
other countries, the ideal FOP system for use with consumers is not clear. Grunert and Wills
(68) have proposed a theoretical model to explain the steps to successful food label use that can
be applied to the context of FOP systems (Figure 2-3). According to the model, exposure to a
FOP system on the food label is the first pre-requite for use of a FOP system in decision-making,
and the chance of exposure is increased if consumers actually search for the system on the label.
If the FOP system is to be used, perception (a.k.a. noticing) of the system and its information
must result from the exposure. What follows perception is liking, or preferences, for the label
and understanding of the label, which can both impact on the use of a FOP system. Grunert and
Wills (68) hypothesize that three basic considerations guide consumers liking of FOP systems: 1)
simplicity; 2) knowledge of what the system stands for and how it was arrived at; and, 3) the
extent to which they feel coerced. Understanding of a FOP system can be of two types:
29
conceptual (objective) understanding and substantive (subjective) understanding (68,134,135).
Conceptual understanding refers to a consumer’s ability to understand the intended meaning
behind a FOP system, while substantive understanding refers to a consumer’s ability to interpret
the information on the FOP system and the extent to which they believe they have interpreted it
correctly. Most of the steps in this model will be influenced by a number of factors, including
the consumer’s interest in, and knowledge of, nutrition issues, the consumer’s demographic
characteristics, and the FOP systems format (68).
Figure 2-3 Conceptual model of FOP system use (adap ted from Grunert and Wills
(68))
2.6.1.1 Consumers’ preferred FOP system
Only a limited number of studies have examined consumer preferences for different FOP
systems (126). In studies comparing consumer preferences for summary indicator versus
nutrient-specific systems, consumers appear to favour nutrient-specific systems (33,62,63,136-
138). A 2007 survey of an ethnically diverse sample of consumers in New Zealand found that
34% of respondents expressed a preference for a nutrient-specific multiple traffic light system,
compared to 24% who preferred a summary indicator single traffic light system, and 13% who
Search
Exposure
Perception
Liking
Understanding
Subjective
Objective
Use
Influences:
Interest
Knowledge
Demographics
FOP system format
30
preferred a nutrient-specific daily value system (33). Likewise, a survey of a large cohort of
French consumers found that 59% of respondents preferred a multiple traffic light system to four
summary indicator alternatives (138). The respondents also perceived the multiple traffic light
system as more likeable and attractive. In addition, Feunekes et al. (63) found consumers from
four European countries perceived a multiple traffic light system as more likeable than four
summary indicator options, and in yet another similar survey (reported on in the same paper),
found that consumers rated a nutrient-specific daily value system as more likeable than three
summary indicator options. Finally, Belgian consumers preferred a nutrient-specific daily value
system to a summary indicator system in a self-administered questionnaire (137). Similarly, in
their review of both the scientific and grey literature on European consumers’ response to
nutrition information on food labels, Grunert and Wills (68) concluded that summary indicators
were the least liked system. Preferences for summary indicator systems that are common in
Canada, including the Heart and Stroke Foundation’s Health Check™ symbol and common
manufacturer developed systems like Pepsi’s Smart Spot and Kraft’s Sensible Solutions, have
not been evaluated against nutrient-specific systems.
In studies that directly compared the two most common nutrient-specific systems – multiple
traffic light systems and daily value systems, – reports of which system is preferred have been
conflicting, but appear to favour multiple traffic lights (33,137,139,140). A qualitative study
conducted with parents in New Zealand reported that consumers perceived traffic light systems
as preferable to use while shopping than daily value systems (139). Equally, as previously noted
a multiple traffic light system was preferred by more respondents than was a daily value system
in a survey of an ethnically diverse group of New Zealanders (33). Further, an Australian study
found consumers rated a traffic light system higher than a daily value system when asked to rate
the systems along several dimensions of consumer friendliness (140). In contrast are the results
of Moser et al. (137), who conducted two similar surveys with populations in Germany and
Belgium that assessed preferences for a daily value system used in both countries compared to a
multiple traffic light system. While German consumers expressed a strong preference for the
traffic light system, Belgian consumers strongly preferred the daily value system. German
consumers rated the traffic light system higher than the daily value system along consumer
friendliness measures such as utility, and design; Belgian consumers rated the daily value system
higher than the traffic light system on along consumer friendliness measures such as clarity,
31
utility, and attractiveness. Daily value systems currently appear on some food labels in Canada
(92,94), however, traffic light systems do not. Therefore it is unclear if Canadian consumers
would display similar preferences, like those noted above, for unfamiliar traffic light systems
versus more familiar daily value systems.
The evidence suggests that perceived credibility or trustworthiness play a role in consumers’
preferences for FOP systems. Feunekes et al. (63) found that in addition to rating a traffic light
system higher on a liking scale than summary indicator systems, consumers also rated the traffic
light system higher with respect to credibility. Similarly, German consumers rated their
preferred traffic light system higher in trustworthiness than their less preferred daily value
system (137).
The results of Moser et al.’s (137) surveys of German and Belgian consumers made clear that
preferences for FOP systems may vary by country, with respondents from these two European
countries having completely opposing preferences for traffic light versus daily value systems.
The results of Feunekes et al’s (63) survey lend further support to this finding: in their survey of
four European countries a significant interaction between FOP system format (six formats were
tested) and country emerged for consumers’ ratings of the likeability and credibility of the FOP
systems. In addition, another European study that tested eight different FOP calorie labelling
systems with consumers in Germany, the Netherlands, France, and the United Kingdom found
differences in consumers evaluations of different FOP systems, as well as in the order of
preferred FOP systems, between countries (10). In addition to differences in preferences at the
country level, there is also evidence to suggest that preferences for different FOP systems differ
by ethnicity (33,62). Gorton et al. found differences in preferences for nutrient-specific versus
summary indicator systems among certain ethnic groups (33). To our knowledge, consumer
preferences for various FOP systems have not been tested in Canada. Given that country-
specific differences appear in consumers’ preferences for various FOP systems, the collection of
Canadian specific data appears warranted.
Finally, research has shown that consumers perceive FOP systems as a valuable addition to
standardized back-of-pack nutrition labelling (10). However, despite the fact that back-of-pack
nutrition labels are mandatory in many countries (15), including Canada (5), only one study to
date has examined preferences for FOP systems in relation to a mandatory nutrition label.
32
Gorton et al. (33) found that 34% of consumers preferred a traffic light system to the mandatory
back-of-pack nutrition label, compared to 26% who preferred the mandatory label. Further, we
are unaware on any study that has tested consumer preferences for different FOP systems in
situations both with and without a standard back-of-pack nutrition label.
2.6.1.2 Understanding
Self-reported understanding of FOP systems was relatively high in most studies, with little
variation in understanding between systems (63,68,132,134,135,138). For example, consumers
in the French survey perceived most FOP systems as easy to understand, with 93% of consumers
or more responding that a variety of summary indicator systems and a multiple traffic light
systems were easy to understand (138). Consumers in the survey of four European countries
scored multiple traffic light systems as easier to understand than summary indicator systems
(63). Comparisons of self-reported understanding of the two common nutrient-specific systems,
traffic light systems and daily value systems, tended to favour the traffic lights (139,140).
However, while consumers in the German survey, discussed previously, rated a traffic light
system as easier to understand than a daily value system, German survey participants from a
another similar study rated the daily value system as easier to understand than the traffic light
system (137). Differences in reported understanding of various FOP systems appear at the
country level, as well as between consumers with different degrees of nutrition knowledge,
interest in healthy eating, label use, and socioeconomic status (63,134,135,137). Self-reported
understanding, however, should be interpreted with caution as research has shown that found that
self-reported understanding of an FOP system does not necessarily translate into better
performance on tests of substantive understanding (63,132).
2.6.1.2.1 Conceptual Understanding
Conceptual understanding (defined in section 2.6.1) of summary indicator systems has not been
directly tested. However, there is evidence to suggest that many consumers attribute the wrong
meaning to them. An early study of the New Zealand National Heart Foundation Heart Tick
symbol found that 43% of consumers thought that to help prevent heart disease you should only
eat foods with the Heart Tick symbol (120). Further, a study of the long-running Swedish
summary indicator symbol, the Green Keyhole, found at least a minimal understanding of the
symbol among consumers (141). When asked to define its meaning 53% of men and 76% of
33
women could, as a minimum, state that the symbol meant a ‘good food’ or ‘healthy food
product’.
With respect to nutrient-specific systems, qualitative research suggests that conceptual
understanding of traffic light systems may be low in some consumers (132). Consumers who
reported they were unfamiliar with such systems thought that the traffic light colours were a
system design element rather than an interpretation of the foods nutrient levels, or that the colour
was related to the nutrient (e.g. fats are always coloured red). Similarly, some misunderstood
daily value systems, with many consumers having a poor understanding of what the percent daily
value means; some consumers interpreted it as the percentage of nutrient in the product. Further,
a study that directly tested conceptual understanding of these two styles of two nutrient-specific
systems found only moderate levels of conceptual understanding among consumers in the United
Kingdom (135). A study that directly tested conceptual understanding of daily value systems in
six European countries found low levels of conceptual understanding with substantial variation
between countries (134). The conceptual understanding of different FOP system styles in
Canadians has not been tested.
2.6.1.2.2 Substantive Understanding
Studies of consumers’ substantive understanding of FOP systems have been inconsistent in their
findings (33,63,65,132,138). A handful of studies in Europe, the United Kingdom, and Australia
have used experimental designs to test the substantive understanding of summary indicator
systems and nutrient-specific systems with consumers (63-65,132,135). Most of these studies
exposed participants to food product pairs, or sets of three, carrying FOP systems and had
consumers identify the healthier product. Most studies that have tested nutrient-based systems
against one another have found either no differences between traffic light systems and daily
value systems, or that traffic light systems are better at helping consumers identify healthier
choices than are daily value systems (64,65,132). For instance, Kelly et al. (65) found 81% of
participants were able to correctly identify the healthier of two products using a traffic light
system, compared to 64% using a monochromatic daily value system. The only conflicting study
found a daily value system was marginally better than the traffic light system at helping
consumers identify the healthier products (88.0% versus 86.6%) (135).
34
The results of studies that tested a combination of summary indicator systems and nutrient-based
systems have been less consistent. For instance, in one study Feunekes et al. (63) found that
graded summary indicator systems (which gave foods one to five stars or smileys) and multiple
traffic light systems were significantly better than score based summary indicator systems at
helping consumer differentiate products based on healthfulness (P<0.01). Furthermore, the
summary indicator systems in general performed most consistently in helping consumers
differentiate between a range of product categories (P<0.01). However, a later study by the same
researchers found virtually no significant differences between summary indicator systems and a
nutrient-based system in helping consumers differentiate between healthy and less healthy
product variants (63). Furthermore, Borgmeier and Westenhoefer (64) found that nutrient-based
systems were superior to summary indicator systems in helping German consumers identify
healthier choices in pair-wise food comparisons (p<0.01).
Other studies have asked consumers to rate the overall healthiness and nutrient content of single
foods to assess subjective understanding of FOP symbols (33,65,132,135,138). For example,
Mejean et al. exposed consumers to three soup variants without a back-of-pack nutrition label
and asked them to respond true or false to five statements related to healthiness and nutritional
value of these soups (138). Consumers demonstrated higher levels of subjective understanding
of summary indicator systems with 58-63% of respondents scoring 4/5 or 5/5 on the true/false
quiz with these systems compared to 53% of respondents achieving the same result with a
multiple traffic light system. Similarly, Gorton et al. (33) exposed consumers to one of three
FOP systems, or a back-of-pack nutrition label, on a package of crackers and asked them
whether they considered the food ‘healthy’ or ‘not healthy. A summary indicator single traffic
light system resulted in the most correct responses (83%), followed by a multiple traffic light
system (80%), and lastly a daily value system (49%). 54% of consumers responded correctly
using the mandatory nutrition label. Studies that compared substantive understanding of
nutrient-specific systems have found that traffic light systems perform better than daily value
systems (65,132). An Australian study found consumers were more accurate in the identification
of nutrient levels when they used traffic light systems compared to when daily value systems
were used (65). Further, two studies have found that consumers could evaluate the overall
healthiness and nutrient content of a single product with a higher degree of accuracy using either
traffic light or daily value systems (132,135). One of these studies also found that substantive
35
understanding of both traffic light and daily value systems could be increased by the presence of
text describing nutrient levels as “high”, “medium”, or “low” (132).
2.7 Summary
There is myriad nutrition information, both mandatory (such as the NFt) and voluntary (such as
nutrition marketing information) found on Canadian food labels. Evidence suggests that
mandatory nutrition information is both under utilized and poorly understood by consumers
while the large amount of voluntary nutrition information may be confusing and misleading
consumers. FOP nutrition rating systems and symbols are a form of voluntary nutrition
information on food labels that has received a lot of attention in recent years. Since their
introduction in the late 1980s, FOP symbols have proliferated in Canada and internationally,
such that there are now a multitude of symbols, each with their own unique appearance and
underlying nutrient profiling criteria. While the intention of FOP systems is to help consumers
make healthier choices, presently, it is unclear if consumers can rely on FOP symbols to identify
products that meet higher nutritional standards. Despite this, consumers appear to perceive
products with FOP symbols as healthier than those without. As a result, experts worry that the
presence of multiple FOP systems in the marketplace may be confusing and misleading
consumers, prompting calls for a single, standardized system to be used on all food products.
Notwithstanding these calls, the ideal FOP system for use with consumers has not been
identified, and Canadian consumers have not been studied to identify their preferred FOP format.
Moreover, Canadians attitudes toward the adoption of a single, standardized FOP system have
not been examined.
36
Chapter 3
3 Scope and Hypotheses of Thesis
3.1 Scope and Objectives
This thesis was designed to generate food supply and consumer data to support the evolution of
Canadian food labelling policy as it relates to FOP systems. In the current Canadian
environment, there is a lack of specific regulations for FOP systems. Presently, it is unclear if
FOP systems are being used to promote foods that meet higher nutritional standards due to a lack
of information on the nutritional quality of foods with FOP systems compared to those without.
Such evidence is needed to evaluate the efficacy of Canada’s current hands-off approach to FOP
system regulation. Moreover, although a Canadian expert committee has called for the adoption
of a single, standardized FOP system to support consumers in making food choices consistent
with a healthy diet (13), there is no evidence to suggest that Canadian consumers would support
such an initiative. Further there is a lack of Canadian data on consumers’ liking and
understanding of different FOP system options to support the identification of an appropriate
single, standardized system should this avenue be pursued. Thus the overall objectives of this
work were to:
1. Assess the extent to which certain voluntary FOP symbols identify all foods whose
nutritional composition meets the system’s nutrient profiling criteria.
2. Compare the calorie, saturated and trans fat, sodium, and sugar content of foods with
FOP symbols to foods without symbols by system type.
3. Describe Canadian’s attitudes to the regulation of FOP systems.
4. Determine experimentally which FOP system is most liked and understood by Canadians.
3.2 Specific hypotheses
We hypothesized that:
1. More foods would meet the nutrient profiling criteria of certain voluntary FOP systems
than would actually carry these symbols on their food label.
37
2. In the absence of minimum, standardized nutrient profiling criteria for foods carrying
FOP symbols on their food labels, foods with FOP symbols versus those without would
contain similar levels of calories, saturated and trans fat, sodium, and sugar.
3. Canadian consumers would support the adoption of a single, standardized FOP system.
4. Canadian consumers would prefer nutrient-specific FOP systems to summary indicator
systems.
5. Canadians’ conceptual understanding of various FOP systems would be low.
3.3 Preview of Chapters 4-6
Objective 1 was examined in Chapter 4 using a 2010-2011 national database of Canadian
packaged food labels known as the Food Label Information Program (FLIP). In this
investigation, we applied the nutrient profiling criteria of two popular summary indicator
systems, a third-party symbol and a manufacturer developed symbol, to the foods in the FLIP
database to determine the proportion of products that would qualify to carry these systems’
symbols. The proportion of products qualifying to carry one of these two symbols was compared
to the proportion of products actually carrying these symbols.
Objective 2 was examined in Chapter 5 by expanding our work with the FLIP database. In this
study we systematically scanned all products in FLIP for the presence of FOP symbol nutrition
marketing, and categorized the FOP symbols used on the label by type (nutrient-specific systems
based on claims criteria, summary indicator systems, food group information systems) as per the
definitions put forward by the US Institute of Medicine. We then compared the calorie, saturated
and trans fat, sodium, and sugar content of foods with and without this form of nutrition
marketing to determine if FOP symbols are a reliable indicator of products with healthier nutrient
profiles.
Objectives 3 and 4 were examined in Chapter 6 using an online, national consumer survey.
Objective 3 was addressed through survey questions designed to discern Canadians’ attitudes
toward the use and regulation of FOP systems with Canada. Objective 4 was addressed through
a randomized mock package experiment that tested five FOP conditions (no FOP systems, two
summary indicator systems, two nutrient-specific systems) on two mock food packages with or
without the mandatory NFt. Consumer friendliness was examined with respect to perceived
38
liking, helpfulness, credibility, understanding, and influence on purchase decisions using a 5-
point likert-type scale.
Together these three studies provide data on the quality of current FOP nutrition rating system
environment, the effectiveness of Canada’s current policies on FOP systems, and suggest future
directions for FOP systems policy in Canada.
39
Chapter 4
4 Food products qualifying for and carrying front-of-pack symbols: a cross-sectional study examining a manufacturer led and a non-profit organization led program
This manuscript has been published: Emrich TE, Cohen JE, Lou WY, L’Abbe MR. Food
products qualifying for and carrying front-of-pack symbols: a cross-sectional study examining a
manufacturer led and a non-profit organization led program. BMC Public Health [Internet]. 2013
Sept [cited 2014 Jan 6]; 13(846):[about 9 pp.]. Available from:
http://www.biomedcentral.com/1471-2458/13/846
This study addressed objective #1 of my thesis, to:
• assess the extent to which certain voluntary FOP symbols identify all foods whose
nutritional composition meets the system’s nutrient profiling criteria.
Student’s contribution:
The original idea for this study and its design were mine. I was involved in the development of
the FLIP database as part of the L’Abbe lab group, including data collection at grocery retailers,
as well as data entry and validation. I independently reviewed each food label in FLIP for the
presence of FOP systems (using the definitions described in the manuscript) and classified each
FOP system identified by type (with validation from a second independent reviewer). For this
study, I independently classified all FLIP products into the appropriate Health Check™ and
Sensible Solutions™ food categories and applied the nutrient criteria for each of these FOP
systems. I designed the analysis of the data with the help of Dr. Wendy Lou and her graduate
student, Ying Qi, and carried out the analyses independently. I completed the original
interpretation of the data and independently prepared a draft of the manuscript prior to engaging
my co-authors in reviewing and revising the manuscript that was published in BMC Public
Health.
40
4.1 Abstract
Background: Concern has been raised that the coexistence of multiple front-of-pack (FOP)
nutrition rating systems in a marketplace may mislead consumers into believing that a specific
food with a FOP is ‘healthier’ than foods without the symbol. Eleven summary indicator FOP
systems are in use in Canada, including one non-profit developed system, the Heart and Stroke
Foundation’s Health Check™, and ten manufacturer-developed systems, like Kraft’s Sensible
Solutions™. This study evaluated FOP’s potential to mislead consumers by comparing the
number of products qualifying to carry a given FOP symbol to the number of products that
actually carry the symbol.
Methods: The nutritional criteria for the Health Check™ and the Sensible Solutions™ systems
were applied to a 2010–2011 Canadian national database of packaged food products. The
proportion of foods qualifying for a given FOP system was compared to the proportion carrying
the symbol using McNemar’s test.
Results: Criteria were available to categorize 7503 and 3009 of the 10,487 foods in the database
under Health Check™ and Sensible Solutions™, respectively. Overall 45% of the foods
belonging to a Health Check™ category qualified for Health Check™ symbol, while only 7.5%
of the foods carried the symbol. Up to 79.1% of the foods belonging to a Sensible Solutions™,
category qualified for Sensible Solutions’s™ symbol while only 4.1% of the foods carried the
symbol. The level of agreement between products qualifying for and carrying FOP systems was
poor to moderate in the majority of food categories for both systems. More than 75% of the
products in 24 of the 85 Health Check™ subcategories and 9 of 11 Sensible Solution™
categories/subcategories qualified for their respective symbols based on their nutritional
composition.
Conclusions: FOP systems as they are currently applied are not, in most instances, a useful
guide to identifying healthier food products in the supermarket as many more products qualify
for these systems than the number of products actually displaying these symbols on FOP, and the
level of agreement between qualifying and carrying products is poor to moderate. The adoption
of a single, standardized FOP system would assure consumers that all products meeting certain
nutritional standards are designated by the symbol.
41
Keywords: Nutrition labelling, Front-of-pack nutrition rating systems, Nutrient criteria
42
4.2 Background
The World Health Organization has stated “consumers require accurate, standardized and
comprehensible information on the content of food items in order to make healthy choices” (3).
To that end, mandatory nutrition labels have been adopted in more than 20 countries, including
the European Union member states, Mexico, and China, and voluntary nutrition labels have been
adopted in at least 11 more (15). In Canada, regulations mandating nutrition labelling on most
packaged foods were adopted 2003 in response to mounting evidence of the contribution of diet
to chronic disease (5). The Canadian Nutrition Facts table reports the amount of calories, fat,
saturated and trans fat, cholesterol, sodium, carbohydrate, fibre, sugar, protein, vitamin A,
vitamin C, calcium, and iron per serving of a food and is similar to the Nutrition Information and
Nutrition Facts panels used in countries such as the US, United Kingdom, Australia, and New
Zealand. At the same time the Nutrition Facts table was adopted, Canada updated regulations for
the use of nutrient content claims and established rules for the use of diet-related health claims
on food products. Canada is just one of many countries, including Japan, China, Australia and
New Zealand, European Union member states, and the US, permitting some form of nutrient or
health claims on food labels (142).
Not included in Canada’s 2003 regulatory revisions was another form of food label nutrition
information, front-of-pack (FOP) nutrition rating systems and symbols. FOP systems provide
simplified information about the nutritional characteristics of a food and have been in use
internationally since American Heart Association first launched its Heart Guide initiative (1987)
and Sweden’s National Food Administration created its Keyhole symbol (1989) (19). Despite
being used internationally for more than 25 years, few specific regulations are in place governing
their use, although standardized FOP systems are being considered in several countries (15). In
Canada a mandatory FOP system is not presently being considered, and the only regulatory
requirement currently governing the use of FOP systems is that they not be “false, misleading, or
deceptive” (26). To minimize the potential for misrepresentation, the Canadian Food Inspection
Agency has issued additional guidance that FOP systems should not give the impression “that a
single food or brand of food is “healthier” than … other foods not bearing the [FOP symbol]”
(26).
43
Since the introduction of the Heart Guide and Keyhole programs, the number of FOP systems in
the marketplace internationally has multiplied (15,19). Each of these FOP systems has their own
unique symbol and nutritional criteria to identify qualifying products. One hundred fifty-eight
unique FOP systems have been identified in the Canadian marketplace, including 11 summary
indicator systems that use a single symbol on products that meet the system’s criteria (78). Of the
summary indicator systems, there was only one third-party, non-profit developed system, the
Heart and Stroke Foundation of Canada’s Health Check™ symbol (Figure 4-1). The Health
Check™ symbol can be placed on qualifying products from any manufacturer (provided the
manufacturer has paid into the program) (143). The remaining 10 systems were manufacturer or
industry-developed and their symbols were placed exclusively on qualifying products of the
proprietary manufacturer (7). Examples of manufacturer-developed summary indicator systems
in use in Canada include Kraft’s Sensible Solutions™ (Figure 4-1), Lassonde’s Health
Signature®, Old Dutch Foods’ Snack Wise™, Pepsi’s Smart Spot™.
Many summary indicator systems are based on nutrient thresholds that establish maximum levels
for nutrients to limit and minimum levels for nutrients or food components to encourage and
often use different thresholds for different food categories (144,145). The number of food
categories covered by each FOP system ranges from as few as one (Snack Wise™ is only
applied to snacks) to as many as 85 (Health Check™ criteria were available for 85 sub-categories
found within 6 major categories at the time data were collected) (145,146). Appendix A lists the
Health Check™ food categories [see Appendix A]. Sensible Solutions™ has developed criteria
for the most food categories of all the manufacturer-developed summary indicator systems, with
criteria available for eight major food categories and five additional subcategories (94,144). An
Appendix lists the Sensible Solutions™ food categories [see Appendix B]. FOP systems in
Canada included both absolute and relative nutrient thresholds (92,94). Absolute thresholds
establish minimum and maximum levels for nutrients and food components, whereas relative
thresholds are set relative to the nutrient content of an appropriate reference product. In the latter
case, products can qualify for a symbol by being lower in a nutrient like calories, fat, saturated
fat, sugar, or sodium, than the reference; thus, symbols can potentially appear on foods high in
saturated or trans fat, sugar, or sodium but low in the nutrient of interest. The nutrient criteria of
summary indicator systems, and manufacturer-developed criteria in particular, whether absolute
or relative, have been accused of not being stringent enough (56).
44
As FOP systems have multiplied, so too have concerns that these systems may be confusing and
misleading to consumers (13,19,92,94). In the most comprehensive review of FOP systems to
date, the Institute of Medicine concluded that the coexistence of multiple FOP systems, with
different nutritional criteria, make it difficult for consumers to interpret nutrition information and
to compare products (109). Expert reports have raised further concerns about the role of FOP
systems in hindering product comparisons, worrying that nutritious foods not part of a FOP
program may, by default, be perceived as less healthy (92,94). This is of concern, as many foods
may not carry FOP symbols for reasons unrelated to their nutritional value. However, just how
many foods are being excluded from carrying a specific FOP symbol for reasons unrelated to
nutritional composition has not been examined.
This study assesses the proportion of Canadian grocery products that qualify for a Health
Check™ or a Sensible Solutions™ symbol based on their nutritional composition compared to
the proportion of products that actually carry these symbols in order to evaluate the potential of
the current FOP systems to mislead consumers.
Figure 4-1. Front-of-pack symbols evaluated in the present study.
A
B
(A) Heart and Stroke Foundation Health Check™ and (B) Kraft Sensible Solutions™.
45
4.3 Methods The FOP systems examined in this research were the Heart and Stroke Foundation’s Health
Check™ and Kraft’s Sensible Solutions™. Health Check™ was chosen as it is the only non-
profit, third-party summary indicator FOP system identified to date in Canadian reports
(78,92,94). Sensible Solutions™ was chosen because it is the manufacturer-developed FOP
system with nutrient criteria established for the largest number of food categories (94).
The nutrient criteria used to determine if a product qualified for the Health Check™ or Sensible
Solutions™ symbol were obtained from the systems’ proprietors (144,145). The Health Check™
criteria are based on levels of total fat, saturated and trans fat, sodium, carbohydrates, fibre,
sugar, protein, and vitamins and minerals and the presence of fruit and vegetables and whole
grains. Sensible Solutions™ criteria are based on the same nutrients and food components as
Health Check™ (145), but also include criteria related to calories, cholesterol, added sugars,
serving size, and functional nutritional benefits (144). Both FOP systems use threshold criteria
but differ, 1) at what level the thresholds have been set, and, 2) with respect to the application of
relative thresholds. In addition to allowing products to qualify for their symbol by meeting
absolute thresholds, Sensible Solutions™ also allows some products to qualify for their symbol
using relative threshold criteria (provided that it passes a review by Kraft’s Nutrition
Department) (144). Examples of Health Check™ and Sensible Solution™ criteria for crackers
are found in Table 4-1 and the remaining criteria are publicly available online from Health
Check™ (http://www.healthcheck.org) and Kraft (http://www.kraftcanada.com).
Data for this study were drawn from the FLIP, a national database of food label information
developed at the University of Toronto (78). The FLIP includes the food label information from
10,487 national and private label grocery products in 23 food categories collected throughout
2010–2011 from the three largest grocery retailers in Canada (Loblaw Inc, Sobeys Inc, and
Metro Inc) and one major western Canadian grocery retailer (Safeway). As previously described
by Schermel et al. (78), by systematically scanning the grocery store shelves we aimed to collect
every food product with a Nutrition Facts table within each of the 23 categories, including all
available national and private label brands, but excluding seasonal products (e.g. egg nog) and
foods from the natural health section of each store. Food products sold at multiple retailers were
only purchased once and when multiple sizes of a product were available, only one size was
purchased. Information recorded from the food labels into the FLIP database included the
46
product name, nutrition information, and FOP symbols used.
Nutrition information from the Nutrition Facts table and the ingredient list were used to
determine which products qualified for the FOP systems under study. Nutrients such as Vitamin
E, magnesium, potassium, and folate, which are not required in the Nutrition Facts table, as well
as the quantity of whole grains or servings of vegetables and fruit in the product were included in
some of the criteria for both the Health Check™ and Sensible Solutions™ systems as “or”
statements (i.e. products could qualify by being a source of one of these nutrients or food
components or by being a source of another nutrient listed on the Nutrition Facts table)
(144,145). In this study criteria were only applied to those nutrients and food components that
were available from the Nutrition Facts table or ingredient list.
All FLIP products were classified into the appropriate Health Check™ and Sensible Solutions™
food categories by a single coder and the nutrient criteria for each of these FOP systems were
applied. Food categories were verified by a second coder in a random sample of 5% of products
and less than 0.5% of verified products were found to be misclassified. For the Sensible
Solutions™ relative nutrient criteria, the mean calorie, fat, saturated fat, sugar, and sodium
content for each food category/subcategory was calculated to create the reference product used to
determine if a product is lower (25%) in these nutrients. It should be noted the Kraft’s relative
threshold criteria were designed to compare products to a base product (such as the original
product variant) or an appropriate reference product (not publicly identified by the
manufacturer). In the absence of information on the composition of the base or reference
products for all products in FLIP, category reference products were established based on means
for the category or subcategory.
Data were analyzed using SAS software (version 9.3, SAS Institute Inc., Cary, NC, 2011).
McNemar’s test was used to compare paired proportions, specifically testing whether the
proportion of products qualifying for FOP symbols was different from the proportion of products
carrying FOP symbols within the same food category or subcategory. Kappa coefficient was
calculated to measure the agreement between products qualifying for and carrying FOP symbols
(147). The kappa coefficient measures the difference between observed agreement and expected
agreement and lies on a scale of −1 to 1, where 0.0 is considered ‘poor’ agreement, 0.2 ‘slight’,
0.4 ‘fair’�, 0.6 ‘moderate’�, 0.8 ‘substantial’�, and 1.0 ‘almost perfect’ agreement. Statistical
47
significance level was set at p�<�.05, unless stated otherwise.
48
Table 4-1. Nutrient criteria for crackers to qualify for Health Check™ and Sensible Solutions™ symbols
FOP system Health Check™ (145) Sensible Solutions™ (9)
Food Category Crackers/Rusks Cookies & Crackers
Amount of Food Per 20 g serving and per on-pack serving Per serving
Must meet all of the following absolute threshold nutrient criteria:
Calories
Fat
Saturated and trans fat
Sodium
Added sugar
Other
No criteria
≤ 3 g
≤ 2 g + ≤15% of calories (combined)
≤5% of total fat
≤ 480 mg (per 50 g)
No criteria
No criteria
≤ 100 calories
≤ 30% of calories
≤10% of calories (combined)
≤ 290 mg
≤25% of calories
A “source of” Vitamin A, C, E, calcium magnesium, potassium, iron, protein, fibre; or,
Contain at least a half-serving of fruit, vegetable, or a nutritionally significant amount of whole grain; or,
Has a functional nutrition benefit.
Or must meet one of the following relative threshold nutrient criteria:
Calories
Fat
Saturated fat
Sugar
Not applicable* Must be free of, or low in, one of these nutrients, or must have 25% less of one of these in comparison to the base product or an appropriate reference product
Must be reviewed by the Nutrition
49
Sodium Department
* The Health Check™ system does not use relative threshold criteria.
50
4.4 Results Criteria were available to categorize 7503 (71.5%) and 3009 (28.7%) of the 10,487 food
products in FLIP under the Health Check™ and Sensible Solutions™ FOP systems, respectively.
Health Check™ and Sensible Solutions™ did not have nutrient criteria established for the
remaining, unclassified foods. FLIP had food products from 81 of Health Check™’s 85
subcategories [see Appendix A]. No food products were collected from the following
subcategories: Fresh fruit (unpackaged foods were not collected); Vegetarian terrines, spreads, or
pates; Egg substitutes; and, Nut and/or seed bars.
Details on food products qualifying for, and carrying, FOP symbols by product category are
found in Figure 4-2. Overall, 3364 (44.8%) of the food products for which Health Check™
criteria were available met the nutrient criteria required to carry the symbol, while only 560
(7.5%) of the food products actually carried the symbol. Similarly, significantly more food
products qualified for the Health Check™ symbol than carried the symbol in 56 of the 85
program subcategories. Full details of the proportion of products qualifying for, compared to
carrying, the Health Check™ symbol by subcategory, including significant differences, are
attached [see Appendix A]. In most subcategories where significant differences were not
observed, there were either a very low percentage of products qualifying or the database
contained very few products in the subcategory. For Sensible Solutions™, when the absolute
threshold nutrient criteria were used, 737 (24.5%) of products for which Sensible Solutions™
criteria were available met the eligibility criteria for the symbol; in contrast, when the relative
nutrient criteria were used, 2379 (79.1%) of the same products were eligible for the symbol.
Overall, only 122 (4.1%) of the products in a Sensible Solutions™ food category carried the
system’s symbol. Full details by subcategory are available [see Appendix B].
False positives (products that carried a symbol but did not meet the criteria) were rare, with more
than 92% of products carrying a Health Check™ symbol meeting the relevant systems criteria.
False positives were found in 13 Health Check™ subcategories with the majority of false
positives in the ‘Combination foods’ subcategories. In most instances, false positives occurred
because the food failed to meet the nutrient criteria per reference amount (a standard serving size
established for each food category), although they met the nutrient criteria per on-pack serving.
51
No false positives were observed with Sensible Solutions™.
There was substantial agreement (kappa >0.8) between the number of products qualifying for
and carrying the Health Check™ symbol in only four subcategories, ‘dried fruit snacks’,
‘croutons’, ‘canned legumes’, and ‘stuffed pasta’. Appendix A provides the level of agreement
(kappa statistic) for all Health Check™ subcategories. Poor agreement (kappa <0.2) between
qualifying and carrying products was observed in many subcategories: ‘vegetable and fruit’ (11
out of 17 subcategories); ‘grain products’ (13/19), ‘dairy products’ (7/10); ‘meat and alternative’
(14/23); ‘fats and oils’ (3/4); and, ‘combination foods’ (4/12). The highest levels of agreement
between products carrying and qualifying for Sensible Solutions™ were observed when absolute
threshold criteria were applied to ‘refreshment beverages’ and ‘cookies and crackers’, however
even within these categories the level of agreement was poor (kappa <0.2). Appendix B provides
the level of agreement (kappa statistic) for all Sensible Solutions™ categories and subcategories.
The Health Check™ and Sensible Solutions™ nutrient criteria were not equally discriminating
in identifying “healthier” choices across food categories (Figure 4-2 and Appendixes A and B).
In eight of the 18 ‘vegetable and fruit’, seven of the 23 ‘meat and alternative’, and two of the
four ‘oils and fats’ subcategories, more than 75% of the products qualified for the Health
Check™ symbol [see Appendix A]. In contrast, in seven of the 12 ‘combination foods’
subcategories, less than 25% of the products qualified for the Health Check™ symbol. With
respect to the Sensible Solutions™ system, fewer than 25% of the foods in each category
qualified for the symbol when the absolute threshold criteria were used, with the exceptions of
‘convenient meal products’ and ‘100% juice’ [see Appendix B]. However, when the relative
threshold criteria were used, more than 70% of foods in each food category qualified for the
Sensible Solutions™ symbol.
The FLIP database contained 409 products made by Sensible Solutions™’s proprietor, Kraft
Canada. Forty-five (11.0%) of Kraft’s products qualified for their symbol on the basis of the
system’s absolute threshold criteria and 361 (88.3%) of their products qualified on the basis of
the system’s relative threshold criteria. However, only 122 (29.8%) of the Kraft products in the
FLIP actually carried the Sensible Solutions symbol. The lower proportion of Kraft products
carrying the Sensible Solutions™, symbol relative to the number of their products that could
52
qualify based on relative threshold criteria suggests that a significant number of products are
disqualified at the required review phase by Kraft’s Nutrition Department.
53
Figure 4-2. Proportion of food products that qualified for, compared with the proportion of
food products that carried, the different front-of-pack symbols.
(A) Heart and Stroke Foundation Health Check™. Criteria were available to categorize 7503 of the 10,487 food products in the FLIP under the Health Check™ system. (B) Kraft Sensible Solutions™. Criteria were available to categorize 3009 of the 10,487 food products in the FLIP under the Sensible Solutions™ system. † Differ significantly from the proportion of food products carrying a front-of-pack symbol p<0.05.
54
4.5 Discussion The findings from the present study showed that significantly more products met the Health
Check™ and Sensible Solutions™ nutrition rating systems’ definition of ‘healthy/healthier’ (as
described in their respective nutrient criteria) than carried either of these FOP symbols in most
food categories. Past research has found that, given two similar foods, one carrying the Health
Check™ symbol and one without it, 80% of consumers would perceive the product with the
symbol as ‘probably a better choice’ while only 4% of consumers would perceive there was ‘no
real difference’ between the two products (121). Similarly, researchers found that consumers
exposed to a FOP symbol on a mousse cake perceived the cake as healthier than consumers who
were given the same cake without a symbol (p�=�0.004) (122). The magnitude of perceived
differences in the healthiness of foods appears to be influenced by the format of the FOP symbol
(60,63,64,109). Given the large number of products that qualify for, yet do not carry these
symbols, our findings suggest that the two FOP systems under study may give consumers the
erroneous impression that foods carrying the symbols are healthier than a similar product without
these symbols – contrary to the Canadian guidance regarding the use of FOP systems (26).
When absolute threshold nutrient criteria were used, a smaller proportion of products qualified
for Sensible Solutions™ than Health Check™. However, when Sensible Solutions™ relative
threshold nutrient criteria were used, a larger proportion of products qualified for the Sensible
Solutions™ than Health Check™. In fact, the relative threshold criteria appeared poor at
differentiating between healthy and less healthy products, with a large proportion of products
qualifying for the symbol in most food categories when these criteria were applied (Figure 4-2).
Based on our findings, relative threshold nutrient criteria were less able to discriminate between
products based on healthiness. However it should be noted that Kraft designed the relative
nutrient criteria to be applied relative to a base product (e.g. a reduced fat Oreo cookie compared
to a regular Oreo cookie) or matched with an appropriate reference product. Thus the use of
category means as the reference nutrient levels for determining which products qualify based on
relative thresholds is a weakness of this analysis. Furthermore, the secondary assessment by
Kraft’s Nutrition Department of products that qualify based on relative threshold criteria is not
documented, and could not be applied in this study.
With respect to the Health Check™ system, this study found only four subcategories where there
was substantial agreement between the number of products qualifying for and carrying symbol.
55
Considering consumers perceive products with the Health Check™ symbol as healthier than
similar products without the symbol (121), our findings suggest that Health Check™ may be a
useful guide to choosing healthier products for consumers in very few subcategories. Most
subcategories within each of the major Health Check™ categories showed only poor agreement
between products qualifying for and carrying this symbol. However, the consumer has no way to
determine in which food subcategories the Health Check™ symbol identifies most products that
meet the system’s definition of healthy, limiting its utility as a guide to healthier choices.
However, universal implementation of a FOP system like Health Check™ or similar threshold
based system to all products (not just those that have bought into the program) would allow
consumers to better differentiate between healthy and less healthy food choices within all food
categories. Indeed, in their 2011 report on FOP nutrition rating systems, the US Institute of
Medicine recommended that an ideal FOP system should be applied to all grocery products
(109).
Proponents of FOP systems suggest these systems have the potential to encourage product
reformulation by manufacturers to meet their nutrient criteria (19). The few studies that have
examined this issue, including one focused on Health Check™, found that FOP systems
successfully encouraged manufacturers to lower the sodium, saturated and trans fat, and calories
in their products (112-115). However, in 24 of 85 Health Check™ subcategories, greater than
75% of products already met the criteria, suggesting that options for reformulation would be
minimal [see Appendix A]. Similarly, when the Sensible Solutions™ relative nutrient threshold
criteria were applied, more than 70% of products in many food categories qualified for the
systems’ symbol. The results of this study would suggest that, within some product
subcategories, the nutrient criteria of Health Check™ and Sensible Solutions™ (especially the
relative thresholds) should be strengthened if they are to encourage the reformulation of more
food products in a healthful way. Indeed, the Health Check™ program has been continually
adjusting its nutrient criteria to encourage manufacturers to reformulate their products to reduce
the amount of nutrients such as sodium and trans fat in the food supply (145,148,149).
The US Institute of Medicine expert committee recommended in their 2011 report that the model
FOP system should be applied universally and be based on absolute thresholds for saturated and
trans fat, sodium, and sugar for two food categories, individual foods and main dishes/meal
products, to allow for the comparison of foods within and across categories (109). In contrast to
56
this recommendation, the two systems in this study, Health Check™ and Sensible Solutions™,
had different nutrient criteria for up to 85 different subcategories, thereby limiting consumers’
ability to use these FOP systems to compare products across categories. Furthermore, the
nutrient criteria used in both of these systems are based on additional nutrients beyond saturated
and trans fat, sodium, and sugar, despite the Institute of Medicine’s finding that there is
insufficient evidence to suggest that including such nutrients in a FOP system would be useful.
Finally, the expert committee proposed that the model FOP system should take a ranked
approach to nutritional guidance where, after meeting a minimum eligibility threshold, products
could earn and display additional nutritional “points” based on their content of those core three
nutrients. However neither system in this study offered additional ranking interpretation of the
nutritional quality of products, thereby limiting their full ability to inform consumers and to
promote continued product improvements.
In addition to the model FOP system proposed by the Institute of Medicine, a number of single,
standardized (mandatory or voluntary) FOP systems are being proposed or adopted by
governments, experts, and industry groups in countries such as the US, European Union member
states, Australia and New Zealand, and South Korea (15). The proposed systems are largely
nutrient-specific, including nutrients such as saturated fat, sodium, and sugar, and display the
amount per serving or per 100 g on the FOP. In addition, the use of traffic light colours to
identify high (red), moderate (amber), and low (green) amounts of nutrients is under
consideration in a subset of these countries. Summary indicator FOP systems, such as Health
Check™ and Sensible Solutions™, are not presently being considered in any jurisdiction for
universal implementation. If Canada were to consider adopting a single, standardized FOP
system they may want to consider an approach more consistent with what is being proposed
internationally, particularly by the Institute of Medicine. The Institute of Medicine proposes two
important features that are not covered by current FOP systems; 1) they recommend a graded
system with one to three stars or checkmarks awarded depending on nutritional composition;
and, 2) products that don't meet the basal criteria, would carry the FOP symbol with zero stars or
checkmarks.
Strengths of this study include the large number of food categories and subcategories examined,
as well as the inclusion of both a non-profit led and a manufacturer led system. In addition, the
FOP systems under study were quantitatively evaluated within the context of the entire food
57
supply and evaluation was not limited to the products of the FOP system’s proprietary
manufacturer or the products of manufacturers that have bought into the non-profit FOP system.
There are a few limitations to this study. First, the nutritional composition of products was based
on the Nutrition Facts table and data were only available on the 14 core nutrients found in the
nutrition label. Both Health Check™ and Sensible Solution™ included some criteria for
nutrients and food ingredients not included in the Nutrition Facts table, thus the present study
may have underestimated the number of products qualifying for these systems based on the
absence of data on these nutrients and food ingredients. Furthermore, the Nutrition Facts table
does not differentiate between total and added sugar, which is used in the Sensible Solutions™
system. As we were unable to differentiate between total and added sugar we applied this criteria
to total sugar and, as a result, may have underestimated the total number of products that would
qualify for Sensible Solutions™. In addition, reliance on nutrient content values reported in the
Nutrition Facts table instead of values determined through chemical analysis may have decreased
the precision of our results. However, one recent Canadian study of five food categories found
no significant differences between nutrient content values reported in the Nutrition Facts table
compared to values determined through chemical analysis for saturated and trans fat, indicating
that the Nutrition Facts table values are quite precise (150).
4.6 Conclusions Within Canada’s current labelling environment, where FOP systems are not universally applied,
nor subject to specific regulations, substantial agreement between the number of products
qualifying for and carrying symbols was only found in a minority of food categories. As a result,
health professionals should advise their clients that FOP symbols, in their current application,
cannot be reliably used to identify food products that meet higher nutritional standards than other
similar products. Overall, many more products qualified for FOP nutrition rating systems than
carried them, thus supporting concerns that FOP systems could mislead consumers into thinking
that products with a FOP symbol are healthier than those without when this is not actually the
case. Given the proliferation of FOP systems internationally, similar analyses should be
undertaken in other countries to determine the extent to which FOP systems highlight all
products that meet higher nutritional standards. As governments and industry groups implement
single, voluntary FOP systems the extent of uptake should be monitored; voluntary systems may
58
rarely be applied to products with poor nutritional quality, and as demonstrated here, if adoption
is not widespread FOP systems may mislead the consumer if they believe products with symbols
to be healthier than comparable products without an FOP. This analysis suggests consumers may
benefit from a single, standardized FOP symbol that identifies all food products that meet a
common set of nutritional standards – such as those proposed or under consideration in several
countries (15).
59
Chapter 5
5 Front-of-pack symbols are not a reliable indicator of products with healthier nutrient profiles
This manuscript has been submitted to Appetite for publication: Emrich TE, Qi Y, Cohen JE,
Lou WY, L’Abbe MR. Front-of-pack symbols are not a reliable indicator of products with
healthier nutrient profiles. Appetite (revision submitted May 21, 2014).
This study addressed objective #2 of my thesis, to:
• compare the calorie, saturated and trans fat, sodium, and sugar content of foods with FOP
symbols to foods without symbols by system type.
Student contribution:
I developed the original idea and design in collaboration with my supervisor Dr. Mary L’Abbe. I
was involved in the development of the FLIP database as part of the L’Abbe lab group, including
data collection at grocery retailers, as well as data entry and validation. I independently
reviewed each food label in FLIP for the presence of FOP systems (using the definitions
described in the manuscript) and classified each FOP system identified by type (with validation
from a second independent reviewer). I designed the analysis with the help of Dr. Wendy Lou
and her graduate student, Ying Qi, and carried out the analyses independently. I completed the
original interpretation of the data and independently prepared a draft of the manuscript prior to
engaging my co-authors in reviewing and revising the manuscript that was eventually submitted
to Appetite.
60
5.1 Abstract
Front-of-pack (FOP) nutrition rating systems and symbols are a form of nutrition marketing used
on food labels worldwide. In the absence of standardized criteria for their use, it is unclear if
FOP symbols are being used to promote products more nutritious than products without symbols.
OBJECTIVES: To compare the amount of calories, saturated fat, sodium, and sugar in products
with FOP symbols, and different FOP symbol types, to products without symbols. METHODS:
The median calorie, saturated fat, sodium, and sugar content per reference amount of products
with FOP symbols were compared to products without FOP symbols using data from the Food
Label Information Program, a database of 10,487 Canadian packaged food labels. Ten food
categories and 60 subcategories were analyzed. Nutrient content differences were compared
using Wilcoxon rank-sum test; differences greater than 25% were deemed nutritionally relevant.
RESULTS: Products with FOP symbols were not uniformly lower in calories, saturated fat,
sodium, and sugar per reference amount than products without these symbols in any food
category and the majority of subcategories (59/60). None of the different FOP types examined
were used to market products with overall better nutritional profiles than products without this
type of marketing. CONCLUSION: FOP symbols are being used to market foods that are no
more nutritious than foods without this type of marketing. Because FOP symbols may influence
consumer perceptions of products and their purchases, it may be a useful public health strategy to
set minimum nutritional standards for products using FOP symbol marketing.
MeSH key words: Food Labeling, Food Quality, Food Analysis
61
5.2 Introduction
Worldwide, chronic diseases account for 60% of deaths, and unhealthy diet is a preventable risk
factor shared by most chronic diseases (1,151). To reduce chronic disease risk, the World Health
Organization recommends that individuals and populations limit their intake of saturated and
trans fat, cholesterol, and simple and added sugars, while achieving energy balance (1,3). To
help consumers choose foods consistent with these recommendations, the World Health
Organization supports the provision of “accurate, standardized and comprehensible information
on the content of food items” on food packages (3). Indeed, in many countries around the world
standardized, voluntary or mandatory nutrition labels are found on the back-of-pack of some, or
all, prepackaged foods (15). For example, Canada has required the use of a mandatory Nutrition
Facts table (NFt) on most prepackaged foods since 2007 (5). Furthermore, voluntary claims that
describe the level of a nutrient in a food or the relationship between a food and health are also
permitted on products meeting prescribed conditions in many countries (119). For example,
Canada’s Food and Drug Regulations allow for the voluntary use of nutrient-content claims such
as “low in fat” and health claims such as “a healthy diet with adequate calcium and vitamin D,
and regular physical activity, help to achieve strong bones and may reduce the risk of
osteoporosis” on food labels. Besides nutrition labels and claims, a variety of front-of-pack
(FOP) nutrition rating systems and symbols have been providing simplified nutrition information
to consumers on the front of food packages since the 1980s (15,19). Standardized, voluntary
FOP systems have been introduced in some countries (16,17); however, multiple FOP systems
with their own unique symbols and underlying criteria can currently be found in most
marketplaces (19,56,106). Within Canada, there are presently no specific regulations governing
the use of FOP symbols, beyond that they may not be “false, misleading, or deceptive” (73).
Voluntary claims and FOP systems provide nutrition information beyond what is required on the
nutrition label in most jurisdictions and can therefore be defined as forms of nutrition marketing
(70).
The US Institute of Medicine has categorized FOP systems into general three types: nutrient-
specific systems, summary indicator systems, and food group information systems (19).
According to the Institute of Medicine, nutrient-specific systems typically either display the
amount of calories and select nutrients per serving (i.e. repeat some of the information required
by nutrition labels on the FOP) or use symbols based on claim criteria (i.e. ‘low in fat’ or ‘high
62
in fibre’). Summary indicator systems provide summary information on the nutrient content of a
food product using a single symbol, icon, or score and are based on nutrient thresholds or
algorithms. Finally, food group information systems use symbols to convey the presence of a
food group or ingredient (see Figure 5-1 for examples of each type of FOP system).
Nutrition marketing has the potential to influence consumer purchases at the grocery store, which
may impact consumption patterns and ultimately chronic disease risk. Consumers perceive
products with summary indicator systems (such as the Heart and Stroke Foundations’ Health
Check™ symbol shown in Figure 1) as more healthful and lower in ‘negative’ nutrients (121-
123). Moreover, it has been found that FOP claims (such as the nutrient-content and health
claims that form the basis of some nutrient-specific systems such as the General Mills’ Goodness
Corner found in Figure 1) exert a ‘halo’ effect whereby consumers tend to generalize the claim
to the entire product, believing the product is healthier with respect to other nutritional and health
elements not identified in the claim (82,83,89). While we are not aware of any study examining
consumers perceptions of products with food group information systems, it is possible that this
“halo” effect may extend to such systems given their similarities with nutrient-specific systems
based on claims criteria. While there is little research available examining the impact of the
different FOP systems on food purchases and consumption (126), 23% of consumers report
looking for better choice slogans, symbols or logos [FOP systems] on food labels (6), and
qualitative research has found that many consumers use FOP nutrition information more often
than back-of-pack nutrition labels (81).
Despite evidence that consumers perceive products with FOP systems as healthier or having
more favourable nutrient contents, it is presently not known if FOP systems are being used to
market products with overall better nutrient compositions. Most countries allow products to
carry claims (like those that form the basis of some nutrient-specific systems) without
considering their overall nutrient composition (119). Furthermore, while summary indicator
systems typically consider multiple nutrients in their criteria, the nutrients included and their
thresholds or algorithms vary from one system to the next (19,106), and food group information
systems typically only consider a single food group or ingredient and not overall nutrient content
(19). Considering the weaknesses in the present voluntary FOP systems, the Institute of
Medicine has suggested that to best promote health, FOP systems need to consider calories and
multiple nutrients, namely saturated fat, trans fat, sodium, and added sugar as these nutrients are
63
of greatest relevance to public health and chronic disease risk (109). However, since FOP
systems only currently consider selected nutrients and food components, they may be used to
market products that are no healthier in their content of these nutrients than products without
such marketing.
In the absence of standardized underlying nutrient criteria, we hypothesize that products
marketed with FOP symbols will provide similar levels of calories, saturated fat, trans fat,
sodium, and sugar to products without symbols. This study aimed to evaluate if FOP symbols
are being used on foods lower in calories, saturated fat, trans fat, sodium, and sugar than foods
without symbols. This study also compared foods with and without different types of FOP
symbols to determine which FOP types were most likely to identify products with more
favourable levels of the nutrients related to health risks.
Figure 5-1: Examples of different front- of
Nutrient Specific Systems
Nutrient-specific systems that display the amount of calories and select nutrients per serving
The Grocery Manufacturers Association and the Food Marketing Institute’s Facts Up Front
Nutrient- specific systems based on claim criteria
General Mills’ Goodness Corner
of-pack symbol types
Nutrient Specific Systems
the amount of calories
The Grocery Manufacturers Association and the Food
specific systems based on claim criteria
Summary Indicator Systems
Heart and Stroke Foundation of Canada’s Health Check™
Kraft’s Sensible Solutions™
64
Heart and Stroke Foundation of
Food Group Information
Systems
Whole Grain Council’s Whole Grain Stamp
65
5.3 Methods
A cross-sectional comparison of the calorie, saturated and trans fat, sodium, and sugar content of
foods with and without FOP symbols was completed using the Food Label Information Program
(FLIP). The FLIP is a Canadian database of food label information developed at the University
of Toronto that has quantified the use of nutrition marketing, such as FOP systems, nutrient
content claims, and health claims, on food labels and collected information on the nutritional
compositions of foods from the NFt (which includes information on calories and 13 core
nutrients such as fat, total carbohydrates, protein, sodium, and sugar) (78). Data were collected
in 2010-2011 and included information on 10,487 unique products. Data were sampled from the
four major Canadian grocery retailers that together accounted for 56% of the grocery sales in
Canada (152). Every product with a NFt available from national and private label brands was
purchased by systematically scanning the grocery store shelves in each aisle and collecting each
unique product with a NFt. Each product was purchased once and only in a single size.
FLIP data were collected for 23 pre-defined food categories and 153 subcategories, as described
in Schedule M of Canada’s Food and Drug Regulations [B.01.001] (26). Schedule M has
established reference serving sizes for each subcategory and is the basis of the criteria for
making nutrition and health claims. Nutrition information for each product was taken from the
NFt, which provides information on the amount of calories and 13 core nutrients in a
manufacturer defined serving size. Information from the NFt was used to derive the amount of
calories and nutrients per Schedule M reference amount in order to standardize serving sizes for
comparison.
Each food label was scanned for FOP systems using the definitions and categories defined by the
Institute of Medicine (19). As previously described by Schermel et al. (78), FOP systems in
FLIP were independently classified by two reviewers, with any disagreements being resolved in
consultation with the research team. This analysis included FOP systems that provided nutrition
information beyond what is required by the NFt (i.e. nutrition marketing): nutrient-specific
systems based on claim criteria, summary indicator systems, and food group information
systems. Nutrient-specific systems that displayed the amount of calories and select nutrients on
the FOP were excluded as they simply repeated NFt required information.
66
5.3.1 Data analysis
This analysis focused on the ten food categories that had the largest number of foods with FOP
systems: 1) bakery products, 2) cereals and other grain products (hereafter referred to as cereals
and grains), 3) combination dishes, 4) dairy products and substitutes (dairy products), 5) fats and
oils, 6) fruits and fruit juices, 7) meat, poultry, their products and substitutes (meat and poultry),
8) snacks, 9) soups, and 10) vegetables. A total of 92 subcategories were found within these ten
food categories. For example, the bakery products subcategory included subcategories such as
‘bread’, ‘bagels, tea biscuits, scones, rolls, buns, croissants, tortillas, soft bread sticks, soft
pretzels and corn bread’, and ‘crackers, hard bread sticks and melba toast’. Data were analyzed
using SAS software (version 9.3, SAS Institute Inc., Cary, NC, 2011). The distributions of all
continuous variables were first assessed and then differences in the amount of calories, saturated
and trans fat, sodium, and sugar per reference amount of foods, with and without FOP symbols,
were compared using Wilcoxon rank-sum test. Statistical significance level was set at p<0.05.
Differences in calorie or nutrient content per reference amount greater than 25% were deemed to
be nutritionally relevant as per the Food and Drug Regulations’ criteria for “reduced” or “lower”
claims (5). Under the Canadian Food and Drug Regulations, in order to carry a claim of being
“reduced” or “lower” (for example “Sodium reduced” or “Lower in saturated fat”), a food must
contain 25% less of the nutrient of interest per reference amount of the food than the reference
amount of a similar reference food. Data are reported as median followed by interquartile range.
Only statistically significant and nutritionally relevant differences are reported in the results and
tables. Additional sub-group analyses were conducted by subcategory and by FOP type
(nutrient-specific system based on claims criteria, summary indicator system, and food group
information system).
5.4 Results
Overall FOP symbols (excluding nutrient-specific systems that only displayed the amount of
select nutrients on the FOP) were found on 17.8% of the products in the FLIP (with some
products having two or more symbols). Nutrient-specific symbols based on claim criteria were
found on 3.4% of the products and, as reported by Schermel et al. (78), 7.5% of the products
carried a summary indicator system, 3.5% a food group information system, and 7.0% a hybrid
system that combined features of two of more of the FOP types. Of the 92 subcategories
67
examined, 60 contained products with FOP symbols. Trans fat was excluded from the results as
no differences >0.1 g were observed in any category.
5.4.1 Differences between products with and without FOP symbols by food category
Products with FOP symbols were not uniformly lower in calories, saturated fat, sodium, and
sugar per reference amount than products without symbols in any food category, and in some
instances were higher in one or more of these nutrients (Table 5-1). Products with FOP symbols
were only significantly lower in calories than products without symbols in the ‘cereals and
grains’ category (209 versus 300 cal), while products with FOP symbols had higher calories than
products without symbols in the ‘soups’ category (120 versus 90 cal) (p<0.05). In contrast,
products with FOP symbols were lower in saturated fat than those without in five categories
(‘bakery products’, ‘combination dishes’, ‘dairy products’, ‘meat and poultry’, and ‘snacks’)
with differences ranging from 0.8-2.3 g (p<0.05). ‘Meat and poultry’ products with FOP
symbols were lower in sodium than products without symbols (356 versus 522 mg) (p<0.05). In
half of the categories (‘cereals and grains, ‘combination dishes’, ‘dairy products’, ‘soups’, and
‘vegetables’), products with FOP symbols were higher in sugar than products without, with
differences as great as 6 g of sugar per reference amount (p<0.05). ‘Cereals and grains’ with
FOP symbols were significantly higher in saturated fat, sodium, and sugar than ‘cereals and
grains’ without symbols.
5.4.2 Differences between products with and without FOP symbols by food subcategory
No differences in calorie, saturated fat, sodium, and sugar content were observed between
products with and without FOP symbols in half of the subcategories examined (30/60) (see
Appendix C). Products with FOP symbols were higher in at least one nutritional component
(calories, saturated fat, sodium, or sugar) of public health relevance than products without
symbols in nine subcategories. For example ‘cookies and graham wafers’ with symbols had
25% more sodium than products without symbols (p<0.05). Products with FOP symbols were
found to be lower than products without symbols in one or two nutrients for 17 subcategories
(p<0.05), saturated fat and sodium were the nutrients most often lower. For example, ‘canned
68
meat and poultry’ with FOP symbol marketing had 0.4 g of saturated fat and 241 mg of sodium
compared to 1.6 g and 413 mg in products without symbols (p<0.5).
Products with FOP symbols were lower in three of the nutritional components analyzed in six
subcategories. For example, ‘coffee cakes, donuts, Danishes, sweet rolls, sweet quick-type
breads and muffins’ with FOP symbols had less calories (133 versus 217 cal), saturated fat (0.8
versus 2.2 g), and sugar (10 versus 17 g) than products without symbols (p<0.05) (Appendix C).
Only ‘meat and poultry with sauce’ with FOP symbol marketing was lower in all four nutritional
components compared to products without symbols: calories (134 versus 266 cal), saturated fat
(0.5 versus 4.2 g), sodium (442 versus 782 mg), and sugar (1 versus 4 g) (p<0.05).
5.4.3 Comparison of different FOP symbol types
None of the different FOP types examined was used to market products with overall better
nutritional profiles than products without this type of marketing (Figure 5-2). Although there
were cases of individual nutrient levels being improved, overall products with and without FOP
symbols had similar nutritional compositions, regardless of symbol type. The largest numbers of
significant differences were observed between products with and without hybrid or summary
indicator symbols (p<0.05). However, no food category with either of these FOP types was
lower in all four nutritional components than products without symbols. For example, while
‘soups’ with summary indicator symbols contained less calories, saturated fat, and sodium than
‘soups’ without symbols (by 74 cal, 1.5 g, and 214 mg, respectively), they were higher in sugar
by 14 g (p<0.05).
69
Table 5-1: Comparison of the nutrient content per r eference amount of products with and without front- of-pack
nutrition rating symbols *
Food Category FOP status
N Calories (kcal) Saturated fat (g) Sodium (mg) Sugar (g)
Bakery Products No 1254 144 [120, 183] 1.2 [0.3, 3.4] 169 [91, 250] 6 [1, 13]
Yes 382 126 [91, 140] 0.5 [0.3, 1.3]† 148 [92, 205] 2 [1, 8]†
Cereals and Other Grain Products
No 548 300 [164, 302] 0.2 [0.0, 0.3]† 1 [0, 57]† 2 [1, 4]†
Yes 229 209 [156, 300]† 0.3 [0.2, 0.5] 109 [2, 200] 6 [3,11]
Combination dishes No 852 300 [230, 340] 3.0 [1.3, 4.7] 709 [509, 880] 4 [2, 7]†
Yes 192 261 [212, 295] 1.5 [0.7, 2.5]† 561 [468, 660] 5 [3, 9]
Dairy Products and Substitutes
No 741 109 [81, 130] 3.5 [1.5, 6.0] 160 [100, 220] 1 [0, 9]†
Yes 98 110 [84, 132] 1.5 [0.4, 3.2]† 126 [100, 210] 7 [1, 14]
Fats and Oils No 395 80 [70, 100] 1.0 [0.8, 2.0] 120 [0, 300] 0 [0, 2]
Yes 81 70 [35, 80] 1.0 [0.3, 1.0] 135 [70, 260] 0 [0, 2]
Fruits and Fruit Juices No 611 120 [100, 130] 0.0 [0.0, 0.0] 10 [0, 25]† 25 [21, 29]
Yes 189 120 [86, 130] 0.0 [0.0, 0.0] 20 [6, 35] 25 [19, 28]
Meat, Poultry, Their Products and Substitutes
No 557 147 [90, 223] 2.9 [1.1, 4.9] 522 [425, 640] 1 [0, 1]
Yes 86 127 [91, 154] 0.6 [0.4, 1.5]† 356 [256, 437]† 1 [0, 1]
Snacks No 384 260 [230, 270] 1.5 [1.0, 3.0] 288 [151, 390] 2 [0, 3]
Yes 87 214 [179, 240] 0.7 [0.0, 1.3]† 233 [74, 360] 2 [0, 4]
Soups No 244 90 [60, 150]† 0.5 [0.0, 1.5] 740 [650, 898] 2 [1, 4] †
Yes 90 120 [90, 150] 0.5 [0.2, 1.0] 625 [480, 650] 4 [2, 7]
70
Vegetables No 489 25 [16, 40] 0.0 [0.0, 0.0] 180 [34, 290] 2 [0, 4] †
Yes 134 30 [20, 50] 0.0 [0.0, 0.0] 71 [15, 290]† 3 [1, 5]
*Front-of-pack nutrition rating symbol nutrition marketing includes: 1) nutrient-specific symbols based on claim criteria, 2) summary
indicator symbols, 3) food group information symbols, and, 4) hybrid symbols; and, excludes nutrient-specific systems that display the
amount of calories and select nutrients per serving. All data are presented as Median and Interquartile range [Q1, Q3]. Calorie and
nutrient amounts are expressed per reference amount and rounded to the number of decimal places provided in the Nutrition Facts table.
Reference amounts are reference serving size amounts found in Schedule M of Canada’s Food and Drug Regulations and are the basis of
the criteria for making nutrient content and health claims in Canada.
† Statistically significant (p<0.05) and nutritionally relevant (≥25%) difference between products with and without a front-of-pack symbol
in the amount of calories or nutrient of interest.
Full details for each subcategory are provided in Appendix C.
71
Figure 5-2: Difference in calorie, saturated fat, sodium, and s ugar content per
reference amount between products with and without front-of-pack symbols by
symbol type
*Statistically significant (p<0.05) and nutritionally relevant (≥25%) difference in calorie or
nutrient content per reference amount between products with and without FOP symbol. Products
without FOP symbols were the reference. Negative percentages indicate instances and amounts
where products with FOP symbols were lower in calories or nutrients than the reference.
Positive percentages indicate instances and amounts where products with FOP symbols were
higher in calories or nutrients than the reference. The dashed line represents a 25% difference in
calorie or nutrient content.
72
5.5 Discussion
The present study found that, while FOP symbols were being used as nutrition marketing on
17.8% of all products, these symbols were not being used to market products that were overall
lower in calories, saturated fat, sodium, and sugar than their counterparts without FOP symbols,
and in some instances were higher in one or more of these nutrients. These findings are
consistent with those of Colby et al (70), who found that nutrition marketing in the form of food
company symbols [FOP systems] were being used to market products high in saturated fat,
sodium, or sugar. Further, from this analysis it is clear that none of the different FOP types are
promoting healthier food choices when multiple nutrients are examined, despite certain FOP
types, such as summary indicator systems, considering multiple nutrients in their criteria. Thus,
when grocery shoppers are comparing similar products, choosing the product with FOP symbol
marketing is no guarantee of a selection consistent with a diet that reduces chronic disease risk.
For example, although consumers may perceive ‘combination dishes’ with summary indicator
symbols as healthier and lower in negative nutrients (121,123), we found that compared to
products without symbols, ‘combination dishes’ with symbols were lower in saturated fat and
sodium, but higher in sugar, and similar in calorie content.
The finding that FOP symbols are not being used to promote products with lower levels of all the
nutritional components of public health concern is worrisome because of the ‘halo’ effect
associated with FOP nutrition information (82,83). Although a nutrient-specific or food group
information symbol may only state a product is ‘low in fat’ or a ‘source of whole grains’,
because of the ‘halo’ effect, consumers may infer that a product is also lower in calories,
saturated fat, sodium, and sugar based on this information when it is not actually the case. For
example, we found that foods carrying food group information symbols based on claims criteria
were, for the most part, equal or higher than foods without these symbols in calories, saturated
fat, sodium, and sugar and thus were not a good guide to selecting products as part of a diet that
reduces the risk of chronic disease. Although FOP symbols have been found to impact
consumers’ perceptions of product healthiness and nutrient content less when used in
conjunction with the NFt (153), FOP nutrition information often leads consumers to truncate
their search for information and skip reading the standardized nutrition label (83). These results
suggest that requiring products to meet comprehensive minimum standards with respect to their
calorie, saturated fat, sodium, and sugar before being able to use a FOP symbol may be a useful
73
public health strategy to ensure that these symbols are being used to market products consistent
with a health promoting diet. Indeed, the Institute of Medicine has proposed that prior to
qualifying for a FOP symbol, products should meet a minimum standard with respect to
saturated and trans fat, sodium, and added sugar content (109).
One of the proposed benefits of FOP systems is that they may stimulate manufacturers to
reformulate existing products in order to qualify to carry a FOP symbol (19). Research has
found that FOP symbols developed by non-profits have led participating manufacturers to
reformulate products, changing their content of one or more nutrients, in order to qualify for a
symbol (112-115). However, we found that overall products with FOP symbols had similar
levels of calories, saturated fat, sodium, and sugar to products without symbols, regardless of
FOP type. This suggests that manufacturers wanting to use FOP symbol marketing in the present
unstandardized marketplace would have no impetus to reformulate their products to lower their
levels of calories and nutrients of public health concern as, in general, the criteria associated with
the different FOP systems are so diverse many products can qualify for one without
reformulation.
5.5.1 Limitations
There are a few limitations to this study, including the somewhat subjective nature of identifying
FOP symbols on food packages in the FLIP. However, every attempt has been made to
minimize any subjectivity by using pre-established definitions and two independent raters to
identify FOP symbols. Another limitation is the comparison of foods at the category level.
Categories contain a variety of foods and comparison of products with and without symbols may
not always be a true comparison of like products. We have attempted to increase the likelihood
of fair comparison by conducting detailed analysis at the subcategory level where similar
products are grouped (See Appendix C). Finally, nutrient content information was derived from
the food label rather than chemical analysis, however a 2011, Canadian study found the values
reported in the NFt to be a reliable indicator of the actual nutrient content as determined through
chemical analysis, at least for saturated and trans fat (150).
74
5.5.2 Conclusions
Although lower levels of either calories or one or more nutrients of public health concern were
identified in products with FOP symbols in some product categories and subcategories, in
general the results of this study demonstrated the calorie, saturated fat, sodium, and sugar content
of products being marketed with FOP symbols are no better than products without FOP
marketing, regardless of the FOP type, in a marketplace without standardized FOP labelling.
This suggests that such symbols are being applied more as a marketing feature in the interest of
selling products than promoting healthier food choices. If nutrition marketing influences
consumer perceptions of product healthiness and nutrient content, and ultimately their product
purchases, the results of this study suggest that minimum standards should be established
regarding the content of nutrients of public health significance for products with FOP symbols.
This will help ensure that consumers relying solely on FOP symbols to guide their selection of
healthier products, as part of a diet that reduces chronic disease risk are not being misled.
Furthermore, the establishment of minimum standards for products to carry FOP symbols would
incentivize manufacturers to reformulate their products to lower the content of nutrients of public
health significance, should they want to use this form of nutrition marketing.
75
Chapter 6
6 Consumer perceptions of the Nutrition Facts table and Front-of-Pack nutrition rating systems
This manuscript has been published: Emrich TE, Qi Y, Mendoza JE, Lou WY, Cohen JE,
L’Abbe MR. Consumer perceptions of the Nutrition Facts table and Front-of-Pack nutrition
rating systems. Appl Physiol Nutr Metab [Internet]. 2014 [cited 2014 Jan 7]; 39(4):417-424.
Available from: http://www.nrcresearchpress.com/doi/abs/10.1139/apnm-2013-
0304?journalCode=apnm#.UsxrK3lZW44
This study addressed objectives #3 and #4 of my thesis, to:
• describe Canadian’s attitudes to the regulation of FOP systems; and,
• determine experimentally which FOP system is most liked and understood by Canadians.
Student contribution:
I developed the original idea and design for this study in collaboration with my supervisor Dr.
Mary L’Abbe. I drafted the original survey used in the study prior to receiving input on its
design from my supervisor, thesis committee, and study collaborators. I completed the pilot test
of the survey in collaboration with Julio Mendoza, who administered the existing Guelph
Consumer Monitor and the Canadian Consumer Monitor (CCM) online panels. I created the
online version of this survey and administered it to the CCM panel with support from Julio
Mendoza. I designed the analysis for this survey in collaboration with Dr. Mary L’Abbe, Dr.
Wendy Lou, and Ying Qi, and carried out the analyses independently. I completed the original
interpretation of the data and independently prepared a draft of the manuscript prior to engaging
my co-authors in reviewing and revising the manuscript that was published in Applied
Physiology, Nutrition, and Metabolism.
76
6.1 Abstract
Preferences for, and consumer friendliness of, front-of-pack (FOP) nutrition rating systems have
not been studied with a Canadian population and studies comparing systems when accompanied
by mandatory labelling, like Canada’s Nutrition Facts table (NFt) are lacking. The purpose of
this study was to evaluate four FOPs relative to the NFt with respect to consumer friendliness
and their influence on perceptions of the healthiness and nutrient content of food. Canadian
consumers (n=3029) participating in an online survey were randomized to score the consumer
friendliness of one of five FOP conditions with or without an NFt and to score the healthiness
and nutrient content of two foods using the provided label(s). Mean differences in scores were
evaluated with ANCOVA controlling for age, gender, and education with Tukey-Kramer
adjustments for multiple comparisons. The NFt received the highest scores of consumer
friendliness with respect to liking, helpfulness, credibility, and influence on purchase decisions
(p<0.05); however consumers still supported the implementation of a single, standardized FOP
system, with the nutrient-specific systems (a traffic light and daily value FOP) being preferred
and scored as more consumer friendly than summary indicator systems. Without the NFt,
consumer ratings of healthiness and calorie and nutrient content differed by FOP. With the NFt
present, consumers rated healthiness, calorie and nutrient content similarly, except those who
saw the traffic light. Their ratings were influenced by the traffic light’s colours. The introduction
of a single, standard nutrient-specific FOP to supplement the mandatory NFt should be
considered by Canadian policy makers.
Keywords: nutrition labelling, front-of-pack nutrition rating systems and symbols, surveys
77
6.2 Introduction Front-of-pack (FOP) nutrition rating systems and symbols provide simplified information to
consumers on the nutritional characteristics of food products (19). Since being introduced in the
US and Sweden in the late 1980s, FOP systems have proliferated, with numerous symbols
developed by governments, expert groups, health organizations, food manufacturers, and
retailers currently in use (15,19). We recently found that 19% of foods in a database of Canadian
packaged food labels carried 1 or more FOP system (78). Four general types were found:
nutrient- specific systems that show the amount per serving of select nutrients or use symbols
based on the criteria used to make nutrient content or health claims on food packages; summary
indicator systems that use a single symbol or score to provide summary information about the
nutritional quality of a product; food group information systems that indicate the presence of a
food group or food ingredient; or hybrid systems that combine characteristics of 2 or more of the
preceding systems (19).
The proliferation of FOP systems has led to concerns that they may be confusing and misleading
to consumers because of their inconsistent appearance, application, and underlying nutrient
criteria, and this has prompted expert groups to call for a single, standardized FOP system for use
on all prepackaged food (13,109). However, consumer attitudes toward the adoption of a single,
standardized FOP system have not been evaluated. Furthermore, only a limited number of
studies have examined consumer preferences for different FOP systems and FOP systems’
overall consumer friendliness with respect to liking, helpfulness, credibility and trustworthiness,
and understanding (126). Most studies have identified a preference for nutrient-specific systems,
specifically Traffic Lights, over summary indicator systems (33,62,63,136,137,139). However,
results vary by country and ethnic group, suggesting a need for country-specific studies
(33,62,63,137). In Canada, where Traffic Lights are not in use and summary indicator systems
are the predominant FOP style (78), the consumer friendliness of different FOP systems has not
been tested. Moreover, although in most jurisdictions (including Canada) FOP systems usually
appear in conjunction with either a mandatory or a voluntary nutrition label on food packages,
the consumer friendliness of FOP systems has been tested only relative to the standard nutrition
label in a single study (33); the consumer friendliness of different FOP systems has not been
tested in situations with or without the standard nutrition label, such as the Nutrition Facts table
(NFt) used in Canada.
78
To use FOP systems to guide the selection of foods, consumers must understand them. This
understanding can be (i) substantial understanding (ability to interpret the label correctly to
identify healthier foods) or (ii) conceptual understanding (general understanding of the concept
behind an FOP system) (134,135). With respect to substantial understanding, several studies
have found that consumers are able to use FOP systems to identify the healthier of 2 products
(126), whereas a more limited number of studies have found that consumers perceived the same
product as healthier when it carried an FOP system than when it did not (122,123). The latter
studies suggest the potentially misleading nature of FOP systems; however, it is unclear whether
this effect occurs across a range of product categories and different FOP system types, such as
nonprofit compared with manufacturer-developed systems, or information-based compared with
interpretive systems. Conceptual understanding of FOP systems has been less studied. A 2010
study of a nutrient-specific system in 6 European countries found low levels of conceptual
understanding, with substantial variation among countries (134); conceptual understanding of
other FOP systems has not been tested.
This research sought to examine many of these research gaps, to:
• Identify consumers’ preferred FOP system for use on packaged foods based on self-
reported preferences and measures of consumer friendliness (including conceptual
understanding);
• Determine how different FOP systems, with or without the NFt, influence consumers’
perceptions of the healthiness of different types of prepackaged foods;
• Determine consumers’ attitudes towards the use and regulation of FOP systems.
6.3 Materials and Methods
6.3.1 Participants The study was conducted together with the Canadian Consumer Monitor (CCM) panel
(http://consumermonitor.ca/). The CCM is a nationally representative panel of Canadian
consumers between the ages of 20 and 69 years who complete on-line surveys related to food
and nutrition every 8 to 10 weeks (89,154). The CCM panel was recruited between March 2010
and January 2011; 31 223 individuals were contacted via email to join the panel, and 6665
completed the baseline questionnaire. The current study was conducted in July and August 2011.
79
Consent forms were sent to each participant’s email when he or she joined the panel, and a link
to the consent form was provided at the start of every survey. Ethics approval was received from
both the University of Toronto and the University of Guelph research ethics boards.
6.3.2 Survey Design and Procedure For the mock package experiment, each CCM participant was assigned randomly to 1 of the 5
test FOP systems (Figure 6-1). The Health Check and Smart Pick symbols were chosen to be
representative of summary indicator systems. The former represented a third-party nonprofit
health foundation symbol; the latter was designed for this study to be representative of
manufacturer-developed symbols (such as Kraft’s Sensible Solutions or Pepsi’s Smart Spot). The
Multiple Traffic Light and the Daily Value system were chosen to be representative of nutrient-
specific systems: the former, an interpretive system; the latter, an information-based system.
Participants were further randomized to view their assigned FOP system with or without an NFt,
for a total of 10 different treatment conditions. The no-NFt condition was designed to simulate
what consumers see on store shelves if they choose not to turn over the package to view the
mandatory NFt. Each participant viewed his or her assigned FOP system–NFt condition on the
mock packages of both a frozen meal and a breakfast cereal. These categories were chosen
because they are believed to be familiar to most participants, minimizing the potentially
confounding effect of participants learning about a new product (83).
For each treatment condition, dependent variables were measured on a 5-point Likert-type scale.
The consumer-friendliness measures of liking, helpfulness, credibility, understanding, and
influence on purchase intention were evaluated for each FOP system. Participants in the no-FOP
system–no-NFt condition were not asked to respond to the consumer friendliness measures,
whereas participants in the no-FOP system–with-NFt condition were asked to respond to the
consumer friendliness measures in relation to the NFt. Additionally, all respondents were asked
to rate their perceptions of the healthiness, as well as the calorie and nutrient content, of both the
frozen meal and the breakfast cereal. The consumer friendliness and perceived healthiness
measures were modelled after those used by Feunekes et al. (63) to test the effectiveness of
different FOP nutrition labelling formats in Europe.
Conceptual understanding of each of the FOP systems was further assessed by asking consumers
to rate, on a 5-point Likert-type scale, a series of statements describing the FOP systems (e.g.,
80
“This symbol is a guide to whether or not the food contains a little or a lot of a certain nutrient”,
“This symbol indicates that the food meets a certain nutritional standard”, etc., with anchors 1
(“does not describe at all”) and 5 (“describes completely”).
Finally, panelists completed additional questions about their preferences regarding the regulation
and use of FOP systems by food manufacturers in Canada. Respondents’ demographic variables
(age, gender, education, province, income) were collected during the recruitment process and the
first survey and were merged with this survey data. The survey was translated, face validated by
nutrition-labelling professionals, reviewed by a plain- language expert, and pilot tested with an
online consumer panel from Guelph, Ontario, prior to being administered to the CCM panel
(155).
6.3.3 Statistical Analyses All data analyses were carried out using SAS version 9.3 (SAS Institute, Cary, N.C., USA). Data
are presented as means (± SD) for continuous variables and count followed by percentage for
categorical variables. Analysis of covariance (ANCOVA) was conducted to compare the mean
rating differences in consumer friendliness, healthiness measures, calorie and nutrient content of
the products, and the importance of including calorie counts and different nutrients in the FOP
systems. The Tukey–Kramer method was used for multiple comparisons adjustment throughout
the study. The mean ratings of statements describing the FOP systems were compared using
ANCOVA with control of variables for age, gender, and education. For all the analyses, p < 0.05
was considered as statistically significant unless stated otherwise.
81
Figure 6-1. Front-of-Pack Nutrition Rating Systems evaluated in the randomized
mock package experiment.
Summary indicator systems:� (A) Health Check, a nonprofit-developed system (The Health Check logo, Health Check word mark, and the Heart and Stroke Foundation word mark are trademarks of the Heart and Stroke Foundation of Canada used under license), and (B) Smart Pick, modelled after a manufacturer- sponsored system. Nutrient-specific systems: (C) The Multiple Traffic Light, an interpretive system, and (D) The Daily Value system, an information-based system. B, C, and D were created by On Brand Design under contract with Dr. Mary L’Abbé at the Department of Nutritional Sciences, University of Toronto. In the fifth FOP condition, participants were exposed to mock packages without an FOP system. Participants were randomized to evaluate an assigned FOP system with or without a Nutrition Facts table, for a total of 10 treatment groups.
A
B C
D
82
6.4 Results
6.4.1 Participants Three thousand twenty-nine respondents (65% women) completed the survey and were included
in the analysis. Demographic characteristics of respondents are shown in Table 6-1. There were
no significant differences in the 10 conditions with respect to gender, age, or education among
the participants. The sample was older and better educated than the general Canadian population.
6.4.2 Consumer friendliness of FOP systems Mean scores of consumer friendliness differed by FOP system and by NFt (with the exception of
understanding, which did not differ by NFt) (Table 6-2). The NFt–no-FOP system label
combination received significantly higher mean scores on liking, helpfulness, credibility, and
influence on purchase decisions than all the FOP system–NFt combinations, except for the Daily
Value system–no-NFt combination, which received similar scores for liking (on both the frozen
meal and the breakfast cereal) as well as helpfulness and credibility (on the frozen meal only).
The 2 nutrient-specific systems (with or without the NFt) received significantly higher scores on
understanding than did the 2 summary indicator systems (with or without the NFt) on both the
frozen meal and the breakfast cereal. In addition, the 2 nutrient- specific systems (with or
without the NFt) received higher scores on liking, helpfulness, credibility, and influence on
purchase decisions than did Health Check (with the NFt) and Smart Pick (with or without the
NFt) (p < 0.05).
6.4.3 Effect of NFt on consumer friendliness of FOPs Pairwise comparison between the frozen meals with or without the NFt showed that, consumers
found the summary indicator systems (Health Check™ and Smart Pick) more helpful, credible,
and likeable when they were shown without the NFt (p<0.05) (Table 6-2A). In addition, the
Daily Value FOP was given significantly higher scores on helpfulness when shown without the
NFt. However, no significant differences in consumer friendliness scores were seen between the
FOP pairs (FOP system with and without NFt) on the breakfast cereal (Table 6-2B).
6.4.4 Conceptual Understanding With respect to conceptual understanding, when presented with a list of possible descriptors of
the NFt or their FOP system, consumers scored the correct descriptions of the concept behind the
83
NFt or specific FOP system the highest (Table 6-3). More than 70% of consumers randomly
assigned to either the NFt or the Daily Value system believed that their respective labels could be
used to compare foods both within (e.g., cereal to cereal) and across (e.g., cereal to crackers)
food categories. Seventy-one percent of consumers in the Traffic Light group believed that it
could help them compare foods within a food category. Fewer than 50% of the consumers in the
Health Check or Smart Pick groups believed these FOP systems could be used to compare foods
within or across food categories.
6.4.5 Effect of FOP on perceived overall healthiness and nutrient content
Mean scores of healthiness and nutrient content differed by NFt and FOP system (Table 6-4).
With respect to the frozen meal, consumers who saw the NFt, regardless of FOP system, gave
similar scores on healthiness and on calorie, fat, saturated fat, trans fat, fibre, and sugar content.
However, consumers who saw the Traffic Light and an NFt scored the frozen meal significantly
lower in sodium than did all the other FOP system–NFt groups.
When there was no NFt, consumers in the Health Check group scored the frozen meal as
significantly healthier than did the 2 nutrient-specific system groups and the no-FOP system
group. In addition, the Health Check group scored the meal lower in saturated and trans fat than
did all the other FOP system groups, but higher in sugar than did the nutrient-specific system and
no-FOP system groups. Consumers in the Health Check and Smart Pick groups without the NFt
scored the frozen meal as healthier, lower in saturated fat and sodium, and higher in fibre, but
also higher in calories and sugar, when compared with the same FOP system with the NFt. In
addition, consumers in the Health Check–no-NFt group rated the frozen meal as lower in fat and
trans fat than did the Health Check–NFt group.
With respect to the breakfast cereals, consumers who saw the NFt with no FOP system, Health
Check, Smart Pick, or Daily Value system scored the cereal similarly with respect to healthiness
and calorie, fat, saturated fat, trans fat, sodium, fibre, and sugar content (Table 6-4). However,
consumers who saw the Traffic Light scored the cereal lower with respect to healthiness and
higher with respect to fat and sugar content than did all other FOP system–NFt groups (p < 0.05).
The Traffic Light group also scored the cereal higher in sodium than did consumers who saw no
FOP system or a summary indicator system (p < 0.05). On the other hand, when the NFt was
84
absent, a number of differences in product ratings among the different FOP systems were seen.
Consumers with the Daily Value FOP system perceived the cereal as significantly healthier than
did consumers who saw the same cereal with no FOP system or Traffic Light. In addition,
consumers who saw the Daily Value FOP system scored the cereal as lower in fat and sugar than
did all other FOP system groups, and lower in sodium than did all other FOP system groups
except for the Health Check group. Consumers scored the cereal as lower in calories and
saturated fat when they saw a nutrient-specific system than did all other FOP system groups (p <
0.05). Finally, consumers who saw either 1 of the summary indicator systems or no FOP system,
without the NFt, scored the cereal as significantly higher in calories and all 7 nutrients than did
consumers exposed to the same FOP system with the NFt. Meanwhile, consumers who saw the
Traffic Light scored the cereal lower in trans fat, sodium, and sugar when the NFt was present
than when it was absent (p < 0.05).
6.4.6 Consumer preferences for FOP systems The majority of participants (86%) believed that there should be either a single mandatory or a
single voluntary FOP system used by all manufacturers, and most (68%) believed that a single
FOP system should be used on all food packages, not just on healthy products or on products
chosen at the manufacturer’s discretion. Participants were split as to who should be responsible
for overseeing the use of FOP systems, with 44% and 35% supporting oversight by the
government and the nonprofit sector, respectively; only 14% favoured oversight by
manufacturers or industry associations. With respect to the importance of including calories and
11 different nutrients in the FOP system, participants rated, on average, all nutrients as at least
somewhat important (mean rating >3.8 out of 5). Sodium, trans fat, saturated fat, and sugar
received the highest mean ratings of importance for inclusion in an FOP system. When asked
directly which FOP system they preferred (Health Check, Smart Pick, Traffic Light, or Daily
Value FOP system), 53% preferred the Traffic Light and 30% preferred the Daily Value FOP
system.
85
Table 6-1. Demographic characteristics of survey re spondents (n=3029)
Characteristics CCM
(N) (%)
Gender
Male
Female
1049
1980
34.6
65.4
Age
20-29
30-39
40-49
50-59
60-69
208
491
756
889
685
6.9
16.2
25.0
29.3
22.6
Education
< high school
High school
Trade school
College
University
32
537
276
1039
1145
1.1
17.7
9.1
34.3
37.7
Province
Alberta
British Columbia
Manitoba
New Brunswick
Newfoundland
Nova Scotia
Ontario
Prince Edward Island
Quebec
Saskatchewan
332
448
189
75
58
167
1078
66
404
172
11.0
16.1
6.2
2.5
1.9
5.5
35.6
2.2
13.3
5.7
Income
< $25,000
$25,000-49,999
182
438
6.0
14.5
86
Characteristics CCM
(N) (%)
$50,000-74,999
$75,000-99,999
$100,000-124,999
$125,000-149,999
$150,000-174,999
$175,000-199,999
$200,000+
Not provided
460
411
317
177
145
83
115
701
15.3
13.6
10.5
5.8
4.8
2.7
3.8
23.1
Note: CCM, Canadian Consumer Monitor.
87
Table 6-2. Mean scores of consumer friendliness by FOP system and NFt* (n=3029)
Part A: Frozen meals
Variables FOP System ANCOVA†
No FOP system‡ (n=600)
Summary Indicator Systems Nutrient-specific systems
Health Check™ (n =
642)
Smart Pick (n = 572)
Traffic light (n = 588)
Daily Value (n = 627)
FOP system
NFt FOP system X NFt
Liking§
No NFt
NFt
N/A
4.1a
3.2de
2.7fg
3.0ef
2.5g
3.5c
3.4cd
3.9ab
3.7bc
<.0001 <.0001 0.0535
Helpfulness||
No NFt
NFt
N/A
4.3a
3.2d
2.8e
2.7e
2.4f
3.6bc
3.3cd
4.1a
3.8b
<.0001 <.0001 0.9561
Credibility¶
No NFt
NFt
N/A
4.0a
3.1c
2.7d
2.4d
2.1e
3.2c
3.0c
3.7ab
3.6b
<.0001 <.0001 0.1086
Understanding**
No NFt
NFt
N/A
4.1bc
3.9cd
3.8de
3.6e
3.6de
4.4a
4.3ab
4.3ab
4.2ab
<.0001 0.0537 0.5808
Influence on purchase decisions††
No NFt
NFt
N/A
4.2a
2.7ef
2.4fg
2.3gh
2.1h
3.1cd
3.0de
3.7b
3.4bc
<.0001
0.0001
0.5327
88
Part B: Breakfast cereal
Variables FOP System ANCOVA†
No FOP system‡ (n=600)
Summary Indicator Systems Nutrient-specific systems
Health Check™ (n =
642)
Smart Pick (n = 572)
Traffic light (n = 588)
Daily Value (n = 627)
FOP system
NFt FOP system X NFt
Liking§
No NFt
NFt
N/A
4.1a
3.1d
3.0de
2.9de
2.7e
3.5c
3.4c
3.9ab
3.7bc
<.0001 0.0017 0.6880
Helpfulness||
No NFt
NFt
N/A
4.3a
3.0d
2.9de
2.6ef
2.5f
3.6c
3.5c
4.0b
3.8bc
<.0001 0.0020 0.5117
Credibility¶
No NFt
NFt
N/A
4.1a
3.1cd
2.9d
2.5e
2.4e
3.3c
3.2cd
3.8b
3.7b
<.0001 0.0182 0.6566
Understanding**
No NFt
NFt
N/A
4.2a
3.8b
3.8b
3.6b
3.6b
4.4a
4.4a
4.3a
4.3a
<.0001 0.5831 0.7284
Influence on purchase decisions††
No NFt
NFt
N/A
4.2a
2.6d
2.5de
2.4de
2.2e
3.2c
3.2c
3.6b
3.5bc
<.0001
0.0127
0.8258
Note: FOP, front-of-pack; NFt, Nutrition Facts table; FOP×NFt, the significance of the interaction between FOP and NFt on consumers ratings of consumer friendliness on the 5-point Likert scale; ANCOVA, analysis of covariance; NA, not applicable.
89
*Values that do not share lowercase letters within each variable are significantly different from each other using LSMeans multiple comparison with Tukey−Kramer adjustments (p < 0.05).
†Models controlled for gender, age, and education.
‡Respondents randomized to view no FOP system and no NFt were not asked to complete consumer friendliness measures; respondents randomized to view no FOP system with an NFt completed consumer friendliness measures with reference to the NFt.
§Question: “How much do you like the [FOP system] on this food?” Anchors: 1 = “do not like at all”, 5 = “like a lot”. Question: “How helpful is the [FOP system] in helping you choose a healthier food?” Anchors: 1 = “not very helpful”, 5 = “extremely helpful”. ¶Question: “How credible is the [FOP system] to you?” Anchors: 1 = “not at all credible”, 5 = “extremely credible”. **Question: “How difficult is it for you to understand the [FOP system]?” Anchors: 1 = “very difficult”, 5 = “very easy”. ††Question: “Would the [FOP system] influence your decision to buy this food?” Anchors: 1 = “not at all”, 5 = “extremely”.
90
Table 6-3. Mean scores of statements describing the NFt and FOP systems
FOP system Statements rated most descriptive of each FOP system Mean Score*
Nutrition Facts table
The Nutrition Facts table is a guide to whether or not the food contains a little or a lot of a certain nutrient.
The Nutrition Facts table is a guide to the amount of different nutrients in a food.
The Nutrition Facts table is a guide to the amount of nutrients a person should eat in a day.
4.0
3.8
2.9
Health Check™ This symbol indicates that the food meets a certain nutritional standard.
This symbol indicates that the food is healthy or good for me.
This symbol indicates that the food is more nutritious than other similar foods.
3.5
3.1
2.9
Smart Pick This symbol indicates that the food meets a certain nutritional standard.
This symbol indicates that the food is healthy or good for me.
This symbol indicates that the food is more nutritious than other similar foods.
2.8
2.7
2.7
Traffic Light This symbol is a guide to whether or not the food contains a little or a lot of a certain nutrient.
This symbol is a guide to the amount of different nutrients in a food.
This symbol indicates that the food is healthy or good for me.
3.5
3.0
2.7
Daily Value This symbol is a guide to whether or not the food contains a little or a lot of a certain nutrient.
This symbol is a guide to the amount of different nutrients in a food.
This symbol indicates that the food is healthy or good for me.
3.6
3.1
2.8
Note: Bolded statements are the correct descriptions of the FOP system of interest. FOP, front-of-pack.
*Anchors: 1 = “Does not describe at all”; 5 = “Describes completely”.
91
Table 6-4. Mean healthiness, calorie and nutrient c ontent scores of a frozen meal
and breakfast cereal by NFt and FOP system *
Characteristics and FOPS
Frozen Meal Breakfast Cereal
No NFt Mean
NFt Mean
P† (NFt*FOP)
No NFt Mean
NFt Mean
P† (NFt*FOP)
Healthiness‡
No FOP system
Health Check™
Smart Pick
Traffic Light
Daily Value
2.5cd
3.2a
3.0ab
2.7bc
2.4cd
2.3d
2.4cd
2.4cd
2.5cd
2.3d
<.0001
3.4cd
3.4bcd
3.6ac
3.2d
3.7ab
3.8a
3.9a
3.8a
3.4cd
3.7ab
0.0028
Calories§
No FOP system
Health Check™
Smart Pick
Traffic Light
Daily Value
4.0a
3.6b
3.6b
2.8d
2.9c
3.1c
3.0cd
2.9cd
2.9cd
3.1c
<.0001
3.4a
3.3a
3.3a
2.2b
2.1b
2.2b
2.1b
2.1b
2.3b
2.2b
0.0028
Fat§
No FOP system
Health Check™
Smart Pick
Traffic Light
Daily Value
3.7a
2.9e
3.2bcd
3.0d
3.0c
3.4ab
3.2bcd
3.3bc
3.2bcd
3.3bcd
<.0001
2.5a
2.2bc
2.4ab
2.2bc
1.7d
1.7d
1.7d
1.6d
2.1c
1.8d
<0.0001
Saturated fat§
No FOP system
Health Check™
Smart Pick
Traffic Light
Daily Value
3.5c
2.4e
2.8d
3.1d
3.7abc
3.9a
3.7abc
3.9ab
3.6bc
3.8abc
<.0001
2.2a
2.0a
2.2a
1.4b
1.3bc
1.2bc
1.2c
1.2c
1.3bc
1.3bc
<0.0001
Trans fat§
No FOP system
Health Check™
3.2a
2.2c
2.9ab
2.7b
0.0001
2.0ab
1.8bc
1.2d
1.2d
<0.0001
92
Smart Pick
Traffic Light
Daily Value
2.7ab
2.7abc
3.0ab
2.9ab
2.7b
2.9ab
2.1a
1.8abc
1.7c
1.2d
1.3d
1.3d
Sodium§
No FOP system
Health Check™
Smart Pick
Traffic Light
Daily Value
4.3a
3.2c
3.7b
3.3c
4.4a
4.4a
4.3a
4.4a
3.8b
4.3a
<.0001
3.0a
2.6bc
2.8ab
2.9ab
2.5c
2.1de
2.0e
2.1de
2.5c
2.3cd
<0.0001
Fibre§
No FOP system
Health Check™
Smart Pick
Traffic Light
Daily Value
2.2b
2.6a
2.6a
2.1b
2.0b
2.2b
2.2b
2.2b
2.1b
2.1b
<.0001
3.8ab
3.8ab
3.9a
3.5ab
3.4b
2.4c
2.3c
2.4c
2.4c
2.4c
0.0036
Sugar§
No FOP system
Health Check™
Smart Pick
Traffic Light
Daily Value
3.1a
2.8ab
2.7b
1.4c
1.2c
1.4c
1.5c
1.4c
1.4c
1.4c
<.0001
3.9a
3.6ab
3.6ab
3.7a
2.7c
2.7c
2.7c
2.7c
3.4b
2.8c
<0.0001
Note: NFt, Nutrition Facts table; FOP, front-of-pack; FOP×NFt, the significance of the interaction between NFt and FOP on consumer ratings of product healthiness and nutrient content.
*Values that do not share letters are significantly different from each other using least-squared means multiple comparison with Tukey−Kramer adjustment (p < 0.05).
†p value for interaction of FOP system and NFt; model controlled gender, age, and education.�‡Question: “How healthy is this food for you?” Anchors: 1 = “not at all healthy”, 5 = “very healthy”.�§Question: “Please indicate if this food has a little or a lot of the following nutrient” Anchors: 1 = “a little”, 5 = “a lot”.
93
6.5 Discussion Overall, we found that the NFt was more consumer friendly than were FOP systems with respect
to liking, helpfulness, credibility, and influence on purchase decisions. This is an encouraging
finding given the widespread use of standardized nutrition labels such as the NFt on food
packages in Canada and internationally (15). However, our findings are in contrast to those of
the only other study to compare a mandatory nutrition label with different FOP systems; Gorton
et al. (33) found that more consumers “liked” or “really liked” Traffic Lights than they did a
mandatory nutrition label. The high ratings of the NFt along the dimensions of consumer
friendliness studied here are perhaps a result of the widespread use of this mandatory label by
Canadian consumers; 71% of Canadians report looking for the NFt when reading food labels,
whereas only 23% report looking for “healthy–better choice” symbols or logos, such as FOP
systems (6). We also found a consumer preference for nutrient-specific over summary indicia-
tor systems, consistent with the findings of other studies (33,63,136). With respect to consumer
friendliness measures, in the majority of FOP system–NFt comparisons, the 2 nutrient-specific
systems (the Daily Value and the Traffic Lights) were scored higher than were the 2 summary
indicator systems (Health Check and Smart Pick) with respect to liking, helpfulness,
understanding, and influence on purchase decisions, and the Daily Value system was scored as
significantly more credible than all other FOP systems.
The widespread use of the NFt by consumers in Canada may also explain the higher levels of
conceptual understanding observed in relation to the NFt. Consumers gave higher scores to the
correct descriptions of the NFt (≥3.8 out of 5) than to the correct descriptions of the FOP
systems, indicating their greater understanding of the NFt. Scores for the correct descriptions of
nutrient-specific FOP systems were similarly high, perhaps because of their greater similarity to
the NFt than to the summary indicator systems.
However, our findings relating to consumer friendliness and conceptual understanding should be
considered in relation to the composition of our sample, which was predominantly female and
better educated than the average consumer. This is particularly relevant given previous research
that has shown that nutrition labels are more likely to be used by women than by men, and by
college-educated consumers than by those with a high school education (156). Furthermore, a
2011 systematic review identified greater understanding of nutrition labels among women and
94
consumers with higher levels of education, literacy, and numeracy; a lack of numeracy skills
appeared particularly problematic in the understanding and use of nutrition labels (157). Thus,
despite our finding that the NFt was superior to the FOP systems along several dimensions of
consumer friendliness and conceptual understanding, other research would suggest that certain
segments of the population may benefit from a more simplified presentation of nutrition
information on food packages in the form of FOP systems. Moreover, CCM panelists strongly
favoured the implementation of a single, standardized FOP system, either by government or a
nonprofit group, despite the higher scores of consumer friendliness and conceptual understanding
attributed to the NFt.
When used in conjunction with the mandatory NFt, FOP systems, with the exception of the
Traffic Lights, had a limited influence on consumers’ perceptions of the healthiness and calorie
and nutrient content of the breakfast cereal and frozen meal used in this study. Compared with
other FOP system groups that saw the NFt, consumers who saw the frozen meal with a “green”
sodium Traffic Light perceived the meal as lower in sodium, whereas consumers who saw the
breakfast cereal with “amber” fat and sodium and “red” sugar Traffic Lights perceived the meal
as higher in fat, sodium, and sugar and as less healthy overall. Thus, it appears that consumers
can correctly interpret the meaning of the Traffic Light colours and incorporate them into their
evaluations of the product.
We found that, as in previous studies, FOP systems influenced consumer perceptions of product
healthiness and nutrient content in the absence of a mandatory nutrition label (122,123).
Andrews et al. (123) found that a frozen dinner displaying a summary indicator system was
rated as healthier overall and lower in negative nutrients than was the same frozen dinner
displaying a Traffic Light nutrient-specific system, or a no-FOP system control. Similarly, we
found that, when the NFt was absent, a frozen dinner with a Health Check summary indicator
system was scored healthier overall and lower in saturated and trans fat than was the same frozen
dinner with no FOP system or a nutrient-specific FOP system. However, a different pattern
emerged when the breakfast cereal was shown without the NFt. Consumers exposed to the
nutrient-specific systems scored the cereal as lower in negative nutrients, demonstrating that the
influence of FOP systems differs by product type.
Finally, with respect to many nutrients, consumers who saw no FOP system or a summary
95
indicator system, without the NFt, scored the frozen meal more favourably, but the breakfast
cereal less favourably, than did consumers who saw the same FOP conditions with the NFt. This
finding is of particular concern because past research has found that the presence of nutrition
information on the FOP label may lead consumers to truncate their search for nutrition
information (i.e., not look at the NFt) and rely exclusively on the claims made on the FOP label
to evaluate products (83).
6.5.1 Conclusion In summary, consumers are in favour of a single, standardized FOP symbol for use on packaged
foods, despite the high levels of consumer friendliness associated with the existing mandatory
nutrition label. Consumers would prefer an FOP system that builds upon the well-liked, helpful,
credible, and influential NFt, providing nutrient-specific information in the form of a Traffic
Light or Daily Value system; however, prior to widespread implementation, such a system will
need to be evaluated as suitable for consumers with lower health-literacy skills. Our findings
suggest that consumers support the recommendation of the IOM that a single, standardized FOP
system appear on all products, and that this should therefore be considered by Canadian policy
makers.
96
Chapter 7
7 Discussion
7.1 General Discussion
Chronic non-communicable diseases are the leading cause of death in Canada and internationally
(23,24). Reducing the burden of chronic disease includes promoting healthy diets as a central
element (1,3,23). Proposed interventions to promote healthy diets have included food labelling
and restrictions on the marketing of unhealthy foods (25). FOP nutrition rating systems and
symbols exist at the intersection of food labelling and marketing and have been identified as a
potential intervention to prevent or reduce obesity and other diet-related chronic diseases (109),
by:
• Increasing the proportion of consumers who use nutrition information to make nutritious
choices for themselves and their families; and,
• Encouraging industry to reformulate their products.
Myriad FOP systems are currently found on packaged food labels in Canada (78). Despite their
pervasiveness within the country, Canada has no specific regulations governing the use of FOP
systems. Further, the present policy of the Canadian Government, not to pursue a single,
standardized FOP system (133), is at odds with the recommendations of experts (13,56,109), as
well as the policies being pursued in a number of other developed nations (15-17).
In 2010, at the outset of this thesis research, Canadian specific data to support the evolution of
Canadian FOP system policies was lacking. As a result, this thesis research explored FOP
systems within the Canadian context with the overarching objective of generating data related to
present and proposed FOP system policies to support the progression of food labelling
interventions in Canada. Three studies were conducted that addressed concerns related to
unregulated, multiple FOP system environments, and proposed single, standardized FOP
systems.
97
7.1.1 Evaluation of existing FOP system policy
The Canadian Government has described FOP symbols as a great source of additional
information for consumers on the foods they purchase (133). Previous investigations have
concluded that the additional information provided by FOP symbols, and summary indicator
symbols in particular, leads consumers to conclude that products with symbols are healthier than
those without (120-124). In Chapter 6, we tested the impact of four FOP systems on perceptions
of the overall healthiness and nutrient content of two foods against a no FOP system control
(153). We found further evidence that the presence of FOP symbols positively influenced
perceptions of healthiness and nutrient content, although the effect varied by FOP system type
and food. For a frozen meal without an NFt, compared to the no FOP system condition,
consumers perceived the food as significantly: healthier, lower in calories, fat, saturated fat, and
sodium, and higher in fibre in the summary indicator conditions; lower in calories, as well as
total fat, saturated fat, sodium, and sugar (nutrients to which the traffic light attributed amber or
green ratings) in the traffic light condition; lower in calories, fat, and sugar in the daily value
condition. For a breakfast cereal without an NFt, compared to the no FOP system condition,
consumers perceived the food as significantly: lower in sodium in the Health Check™ summary
indicator condition; lower in calories, as well as fat and saturated fat (nutrients to which the
traffic light attributed amber and green ratings, respectively) in the traffic light condition;
healthier, lower in calories, fat, saturated fat, trans fat, sodium, and sugar in the daily value
condition. Taken together, consumers clearly attribute more healthy characteristics to foods with
FOP symbols than to those without. These data provide evidence that FOP systems (including
summary indicator systems and daily value systems common in the Canadian marketplace)
contravene current Canadian guidance on the use of third-party FOP symbols described in
section 1.2.2.2 – FOP systems should not “give the impression that a single food or brand is
“healthier” than, or nutritionally superior to, other foods not bearing the … symbol … ” (26).
Prior our study described in Chapter 6, no studies examined the moderating effect of a
mandatory nutrition label, such as the NFt, on consumers’ perceptions of a foods’ healthiness
and nutrient content in the presence of a FOP system. However, Andrews et al. (123) did find
that that the availability of the US Nutrition Facts panel led to more accurate use of nutrition
information by consumers shown a frozen meal with or without a summary indicator system.
Consistent with this, we found that, when presented with an NFt, consumers rated the overall
98
healthiness and nutrient content of a food with a summary indicator system or daily value system
similarly to when no FOP system was present (153). These data suggest that the potentially
misleading impressions made by FOP systems may be attenuated if they are used in conjunction
with a mandatory nutrition label, such as the NFt. However, previous research has found that
FOP nutrition marketing information leads consumers to limit their search for nutrition
information to the FOP such that back-of-pack mandatory nutrition labels may go unused (83).
Therefore, the attenuating benefits of the NFt may be lost if the consumer does not turn the
package over or look for the NFt. Further, our results suggest that traffic light colours may in
fact override the moderating effect of a mandatory nutrition label (153). When the NFt was
given, consumers in the traffic light condition perceived 1) the frozen meal as lower in saturated
fat and sodium (nutrients attributed yellow and green ratings, respectively), and 2) the breakfast
cereal as less healthy, as well as higher in fat, sodium, and sugar (nutrients attributed yellow,
yellow, and red ratings, respectively) than consumers in the no FOP system condition.
In our investigations in Chapters 4 and 5, we used novel approaches to assess the value of FOP
systems in helping consumers make nutritious choices. Although the presence of a FOP symbol
gives consumers the impression that a food is healthier or nutritionally superior to foods without
symbols, prior to our studies there were no investigations that examined whether foods with FOP
symbols were in fact healthier and nutritionally superior to foods without symbols. In Chapter 4,
we compared the proportion of packaged food products that meet the nutrient profiling criteria of
two of Canada’s most common summary indicator systems (Health Check™ and Sensible
Solutions™) to the proportion of packaged food products that actually display these systems’
symbols on their food labels (158). We found that a large proportion of products (44.8-79.1%)
met the nutrient profiling criteria for one of these systems or the other, whereas these systems’
symbols were only found on the food labels of a minority of products (4.1-7.5%). This was the
first study to examine the proportion of packaged food products that meet the nutrient profiling
criteria a FOP system to the proportion of packaged food products that actually display its
symbol. The results of this study show that the use of such FOP systems may in fact mislead
consumers using such symbols to think that they are making healthier choices.
In Chapter 5, we compared the calorie and nutrient content (fat, saturated fat, trans fat, sodium,
and sugar) of foods with an FOP symbol on their label to foods without a symbol. We found that
foods with FOP symbols were not consistently lower in calories, fat, saturated fat, trans fat,
99
sodium, and sugar than foods without symbols. We interpreted the data from Chapters 4 and 5
as evidence that consumers cannot rely on the presence of a FOP symbol as an indicator of foods
that meet higher nutritional standards, despite the impression these symbols may give off.
Furthermore, these data suggest that FOP symbols may be in violation of subsection 5.1 of the
Food and Drugs Act that prohibits the labelling of foods in a way that is misleading or deceptive.
Experts have suggested that the potential for FOP systems to stimulate product reformulation
will contribute to their efficacy in preventing and reducing obesity and diet-related chronic
disease (109). Indeed numerous studies have found a positive effect of third-party summary
indicator systems on stimulating product reformulation among participating manufacturers (112-
115), although studies of third-party nutrient-specific systems have failed to demonstrate a
similar effect (116). However, a necessary pre-requisite for a FOP system to stimulate
reformulation is that the majority, or large proportions, of existing products wishing to carry the
system’s symbol do not already meet the system’s nutrient profiling criteria. The investigation
in Chapter 4 found that, depending on the product sub-category, up to 100% of products already
meet the nutrient profiling criteria of Canada’s only third-party summary indicator system (158)
– leaving little or no impetus for reformulation.
Further, in Canada’s present regulatory environment for FOP systems, manufacturer’s wishing to
market their products with a FOP symbol have the option of creating their own proprietary FOP
system, with their own nutrient profiling criteria, if their products do not qualify for an existing
third-party system, or if they do not wish to participate in a third-party system. This may
account for our Chapter 5 finding that products with FOP symbols have similar levels of calories,
total fat, saturated fat, trans fat, sodium, and sugar to products without symbols. This finding
held true for summary indicator systems, despite the fact that these systems typically consider
multiple nutrients in their nutrient profiling criteria.
In summary, our data provide the first empirical evidence that FOP systems in use in Canada
may be misleading and deceptive, in direct violation of existing guidelines and acts governing
their use. Moreover, Canada’s present regulatory environment for FOP systems does not appear
to support FOP systems’ dual purposes of 1) helping consumers make nutritious choices, and, 2)
encouraging industry to reformulate.
100
7.1.2 Evaluation of potential FOP system policy
In 2011, Canada’s Minister of Health asserted that Canadians have all “the tools they need to
make healthy food choices when they shop for groceries” by way of justifying the federal
government’s position not to pursue a single, standardized FOP system (133). In Chapter 6, we
surveyed Canadian consumers to gauge their support for the implementation of a single,
standardized FOP system (153). Overwhelming support (86%) was found for the adoption of
either a mandatory or voluntary single, standardized FOP system. Furthermore, most consumers
wanted the system to be universally applied to all food products. We interpret this as evidence
that, contrary to the position of the government, Canadians do not believe they have all the tools
they need to make healthy food choices. Indeed, Canadian consumers expressed a desire for a
“standardized system to rate the overall nutritional quality of a food product” in a recent
consultation conducted by Health Canada to support the modernization and improvement of
food labels (159).
Our third investigation, reported in Chapter 6, tested consumer liking and conceptual
understanding of four FOP systems to order to identify Canadians’ favoured FOP system for use
on packaged foods (153). Our survey, conducted in 2011, was the first to explore different FOP
systems with Canadian consumers. Consistent with international research (33,62,63,136-138),
we found nutrient-specific systems were preferred over summary indicator systems on a range of
attributes (153). Another survey of Canadian consumers conducted later in 2011 also found
nutrient-specific systems were rated more highly than summary indicator systems on a range of
attributes (160). In our Chapter 6 investigation, we also found that of the two nutrient-specific
systems, consumers rated the daily value system highest on attributes of consumer friendliness
such as liking, helpfulness, credibility, and influence on purchase decisions (153). However, in a
side-by-side comparison, more consumers in the study selected the traffic light system as their
preferred FOP symbol than selected the daily value system. Our Chapter 6 data suggest that, in
addition to being better liked by consumers, nutrient-specific systems are also better understood.
Nutrient-specific systems received higher ratings of self-reported understanding than did
summary indicator systems. Moreover, a test of conceptual understanding suggested that
nutrient-specific systems are better understood than manufacturer developed summary indicator
systems.
101
Our study, presented in Chapter 6, is the only one to date to examine the impact of a mandatory
nutrition label, the NFt, on how consumers perceive different FOP systems (153). Consumers’
ratings of the likability, helpfulness, and credibility of summary indicator systems were lower in
the presence of the NFt. Similarly, the daily value system was rated less helpful when the NFt
was present. In contrast, consumers rated the traffic light system similarly in both the presence
and absence of the NFt.
In summary, our data provide the first evidence that Canadians desire the adoption of a single
standardized FOP system. Further, the evidence demonstrates that nutrient-specific FOP systems
are Canadians’ preferred FOP system, as they were better liked and understood than summary
indicator systems. These findings are important should Canada pursue the adoption of a single,
standardized FOP system as, according to the theoretical model put forward by Grunert and
Wills (68) (Figure 1-2), liking and understanding of a FOP system are pre-requisites for its use
by consumers.
7.2 Future directions
7.2.1 Policy
Our work provides a foundation to support the development of better policy concerning the use
of FOP systems on food labels in Canada. At a minimum, it suggests the need for expansion and
enforcement of existing guidelines for FOP system use. However, the evidence also makes a
strong case for the development of new policies in this area.
7.2.1.1 Enforcing and expanding existing guidelines
FOP systems currently fall under a range of federal regulations and guidelines depending on the
FOP system category as described in sections 1.2.2 and 1.3. Of particular interest are guidelines
put forth in the Canadian Food Inspections Agency’s Guide to Food Labelling and Advertising
that state third-party [FOP systems] should not give the impression that (26):
1. a single food or brand of food is "healthier" than, or nutritionally superior to, other foods
not bearing the … symbol …
2. the food is a treatment, preventative of or cure for disease.
102
With respect to the first point, we – as well as other researchers – have presented evidence that
FOP symbols (third-party or otherwise) are giving consumers the impression that a single food is
“healthier” or nutritionally superior to other foods (120-124,153). In particular the evidence
most strongly supports the role of summary indicator systems in generating these impressions.
To avoid misleading consumers, the Canadian Food Inspection Agency suggests that in addition
to the FOP symbol, at least one of the following should appear on the food label:
1. A statement that clearly explains the reason for the appearance of the [FOP symbol].
2. The name of the third party [who owns the symbol] … is clearly shown, in conjunction
with its nutrition recommendations or dietary guidelines or those it endorses.
3. A clear indication that the [symbol] … does not constitute an endorsement of the food.
Indeed, the Heart and Stroke Foundation require that participating products carry an explanatory
message on-pack in conjunction with its Health Check™ symbol (the only third-party summary
indicator system in Canada), stating “The Heart and Stroke Foundation’s Registered Dietitians
have reviewed this product to ensure it meets the specific nutrient criteria developed by the
Health Check™ program based on the recommendations in Canada’s Food Guide. A fee is paid
by each participating company to help cover the cost of this voluntary, not-for-profit program”
(161). Manufacturer developed FOP systems are not covered under this guideline, and as a
result most manufacturer-developed systems come without such an explanatory message;
additionally, the criteria they use for their systems are often not readily available. As a result, we
recommend that the Canadian Food Inspection Agency expand the guidelines, currently only
applied to third-party systems, to all summary indicator systems and take action to enforce these
guidelines wherever summary indicator systems are found without additional information
stipulated to in the Guide to Food Labelling and Advertising.
Beyond maintaining the status quo with respect to FOP system policy, the expansion and
enforcement of existing guidelines represents the bare minimum that can be done to reduce the
misleading nature of the current FOP situation in Canada. However, this minimal policy
approach would still be lacking in three important aspects: 1) it would not ensure that consumers
could rely on FOP systems to help them identify healthier choices, 2) it would not provide
manufacturers with any impetus to reformulate, and 3) such a policy would not overcome the
confusion associated with multiple symbols and logos (each relying on different nutritional
103
criteria for their use), identified in previous consumer research (131,132). As a result, we do not
recommend expanding and enforcing existing guidelines as the preferred course of action.
Instead, we would recommend the government pursue one of the courses of action explored in
sections 7.2.1.2 or 7.2.1.3 such as establishing a regulated definition for ‘healthy’ for use in food
labelling as is done in the US, or introducing a single, standardized FOP system.
7.2.1.2 Establishing a definition for ‘healthy’ foods
We have presented evidence that foods with FOP symbols have similar calorie and nutrient
contents to foods without symbols. Further, our data suggest that specific FOP systems,
including third-party and manufacturer developed systems, are not a reliable indicator of foods
that meet a higher nutritional standard. In section 6.1 we presented the dual objectives of FOP
systems in promoting population health: supporting healthier choices by consumers and
reformulation by industry. Our findings suggest that to achieve these purposes the government
should establish a minimum standard for the nutrient profiling criteria behind FOP systems. The
development of minimum standards for nutrient profiling criteria would support two potential
FOP system policy alternatives: 1) the development of a single, standardized FOP system, or 2)
new regulatory requirements that establish prescribed conditions for FOP symbol use – like those
currently in place for nutrient content, disease risk reduction, and therapeutic claims.
A 2009 consultation with stakeholders from organizations with policies or programs supported
by nutrient profiles found that 2/3 of them were in favour of the Canadian government
establishing a standardized nutrient profile to define ‘healthy’ foods (94). Supporters believed
that a standard nutrient profiling criteria to define ‘healthy’ foods would help consumers to adopt
a healthier eating pattern and encourage manufacturers to reformulate or innovate healthier
products consistent with this definition. In the US, nutrient profiling criteria have already been
established for the implied claim implicit in the use of the word “healthy” on food labels (Table
7-1) (19). These criteria would also apply to implied “healthy” claims such as those
communicated by summary indicator systems. Further, these nutrient profiling criteria are also
used in the US in relation to nutrient-content claims. Foods carrying nutrient content claims that
do not meet these minimum criteria must carry disclaimers to alert consumers to negative aspects
of the product – i.e. nutrient levels that exceed the “healthy” threshold.
104
Canada currently has no definition of “healthy” in place. Although, Health Canada recently
introduced nutrient profiling criteria to classify foods from the four food groups into four tiers
based on their compliance with Canada’s Food Guide guidance, Health Canada describes these
criteria as exclusively a surveillance tool (111).
In their 2011 report, Front-of-Package Nutrition Rating Systems and Symbols: Promoting Health
Choices, the US Institute of Medicine (109) issued a series of recommendations for the
development of nutrient profiling criteria that may be edifying to Canadian policy makers. The
Institute of Medicine states that FOP system nutrient criteria should:
1. Target nutrients of public health concern (saturated and trans fat, sodium, and added
sugar).
2. Evaluate nutrient amounts consistent with existing science-based regulations.
3. Apply a single set of nutrient criteria of across all/most product categories (to facilitate
comparisons within and across food categories).
4. Allow compliance to be monitored.
With respect to point #2, the Institute of Medicine suggested that nutrient profiling criteria be set
using the nutrient levels set by the US Food and Drug Administration to define eligibility for
nutrient content and health claims. Similarly, Canada could use the nutrient levels set out in the
Food and Drug Regulations for “low” nutrient content claims to set nutrient profiling criteria.
However, criteria for sugar would need to be developed, as no claim currently exists for “low in
sugar” or “low in added sugar” although levels established in other jurisdictions could be applied
or adapted. Further, with respect to point #4, compliance with added sugar criteria cannot be
monitored through chemical analysis and there is no requirement for the declaration of added
versus total sugars in the NFt. Compliance with added sugar criteria would be difficult to
monitor and require review of the ingredient list. However, a recent Health Canada proposal for
modernizing the food label has proposed the addition of ‘added sugars’ to the NFt, and if
accepted would effectively eliminate this challenge (162). However, nutrient profiling criteria
could be established with respect to total sugar rather than added sugar to facilitate monitoring,
105
as was done in Health Canada’s recently introduced nutrient profiling criteria to classify foods
from the four food groups into tiers (111).
Establishing nutrient profiling criteria to define healthy in the current multiple FOP system
environment would help ensure FOP systems support consumers in identifying healthy foods by
ensuring that foods carrying a FOP system of any kind all meet a consistent standard.
Additionally, standard minimum criteria would incentivize manufacturers to reformulate their
products in order to use FOP systems as nutrition marketing on their food labels. However, as
with the policy option of expanding and enforcing existing guidelines, a policy that supports the
continued use of multiple FOP systems within a single marketplace, albeit with standardized
nutrient criteria, would do nothing to alleviate consumer confusion. The implementation of
some form of standardized FOP system supported by a standard definition of “healthy” is the
policy approach most likely to alleviate consumer confusion.
Table7-1. US Food and Drug Administration and US De partment of Agriculture
nutrient profiling criteria for “healthy” claims (109)
Food Individual Food Seafood/Game Meat Meal or Main Dish
Amount of Food Per reference amount and labeled
serving
Per reference amount and per 100
g
Per 100 g and labeled serving
Fat ≤ 3 g ≤ 5 g ≤ 3 g and ≤ 30% or less of calories
Saturated fat ≤ 1 g and ≤ 15% or less of calories
≤ 2 ≤ 1 g and ≤ 15% or less of calories
Sodium ≤ 480 mg* ≤ 480 mg* ≤ 600 mg
Cholesterol ≤ 60 mg ≤ 95 mg ≤ 90 mg
Beneficial nutrient† ≥ 10% daily value ≥ 10% daily value ≥ 10% daily value of two
Fortification In accordance with fortification policy
* Except for foods with a reference amount ≤ 30 g or 2 Tbsp. must contain ≤ 480 mg sodium per 50 g and ≤ 60 mg cholesterol per 50 g
†Beneficial nutrient could be vitamin A, vitamin C, calcium, iron, protein or fiber
106
7.2.1.3 Implementing a single, standardized FOP system: A discussion of policy options
The results from this thesis provide important evidence regarding a range of potential policy
options that could be applied to the FOP labelling situation. Our research has demonstrated a
strong preference for the implementation of a single, standardized FOP system among Canadian
consumers (153). This is in agreement with the most comprehensive review of FOP systems
completed to date. The Institute of Medicine concluded a model FOP symbol system should be
characterized by a single, standard symbol (109). Taken together, this information suggests
Canadian policy makers should revisit their current position regarding FOP systems.
Although our research indicated consumers prefer a mandatory system to a voluntary system, a
standard, voluntary FOP system might still be considered as a viable policy option. Policy
makers in the European Union have introduced a voluntary daily value FOP system, while
voluntary summary indicator systems are in use in countries such as Denmark, Norway, Sweden,
and the Netherlands (15). Most recently, the United Kingdom and Australia have introduced
voluntary FOP systems for use on packaged foods, with the intent to regulate if voluntary uptake
by industry is limited (16,17). All these programs are underpinned by standardized nutrient
profiling criteria. Similarly, Canada could introduce a voluntary FOP system with specific
nutrient profiling criteria and guidelines for the system’s appearance. This could increase
consistency in the marketplace, and potentially decrease consumer confusion. However, if
voluntary adoption of a standardized system by manufacturers is not widespread the persistence
of competing symbols in the marketplace may lead to continued confusion.
To reduce the confusion generated by multiple competing FOP systems, the Government of
Canada could prohibit the use FOP systems outside the standard voluntary FOP system. Indeed
the Institute of Medicine has recommended that a standardized FOP system be used to the
exclusion of proprietary FOP systems (109). However, a voluntary, exclusive FOP system
policy approach may still be misleading to consumers if they are unclear as to why some foods
carry the FOP system and others to do not.
To truly allow for comparison, the Institute of Medicine has recommended that a standardized
FOP system appear in a consistent location on all food products (109), leading to our final policy
option the implementation a mandatory, standardized FOP system. This approach would be akin
107
to current regulations in place related to the NFt with prescribed formatting and clear guidelines
describing which products must carry the system. A consistent and mandatory FOP system with
a standard format and nutrient profiling criteria would be the system most likely to facilitate the
selection of healthier choices by consumers, as well as stimulate product reformulation among
manufacturers. Further, it would ensure FOP systems are not misleading or deceptive, and avoid
the consumer confusion associated with multiple FOP systems or the inconsistent application of
FOP systems and their nutrient profiling criteria. However, in the current political climate,
where voluntary public health nutrition initiatives have been preferred over regulatory ones in
cases that involved the food industry in recent years, such a policy is unlikely (14,163).
Moreover, legal scholars have raised concerns that an outright ban on the use of proprietary FOP
systems may lead to challenges under the Canadian Charter of Rights and Freedoms (164).
7.2.1.3.1 Format With respect to a standardized FOP system for Canada, our Chapter 6 investigation has provided
evidence that Canadian consumers may be best served by a nutrient-specific FOP system.
Nutrient-specific systems have several strengths relative to the other FOP systems and should be
strongly considered if a single, standardized FOP system is pursued. First, they typically apply a
consistent approach across all or most food and beverage categories, facilitating consumer
product comparisons both within and across food categories (19). In addition, they are consistent
with existing regulations for declaring nutrient amounts per serving and for nutrient content
claims. Further, nutrient-specific systems that display the amount of calories and other nutrients
on the FOP, as in the Daily Value system we tested, allow consumers to make food selections
within the context of their total diet using the percent daily value. Furthermore, nutrient-specific
systems that characterize nutrient amounts as “high”, “medium”, and “low”, as in the traffic light
system we tested, would add further ‘interpretive’ information that helps consumers make food
choices consistent with dietary guidance. From a manufacturer perspective, a nutrient-specific
system may encourage product reformulation to meet the nutrient profiles targeted in the system.
Limitations associated with nutrient-specific systems relate to limited amounts of space available
on labels, which may restrict the amount of information that can be included in an FOP symbol.
Further, there is concern that the inclusion of too many nutrients may clutter the label and
decrease consumers’ ability to use the label.
In addition, the two nutrient-specific systems preferred in our Chapter 6 investigation are
108
consistent with a number of the conclusions reached by the US Institute of Medicine in their
2010 and 2011 reviews of FOP nutrition rating systems and symbols (19,109). Namely, these
systems declare calorie and serving size information, which the Institute of Medicine concluded
was important to facilitate the selection of lower-calorie foods and/or appropriate serving sizes to
combat overweight and obesity (19). Further, both systems highlight most of the nutrients
deemed to be of greatest public health significance –saturated fat, sodium, and sugar (19,109).
Trans fat was omitted from the labels we tested but could be substituted in place of total fat,
which the Institute of Medicine concluded lacked sufficient evidence to support its inclusion
(19). Finally, these nutrient-specific systems would be reinforcing of the existing mandatory
nutrition label, the NFt.
However, these formats diverge from the Institute of Medicine recommendations with respect to
several key points. First, the Institute of Medicine recommended that FOP systems translate
nutrition label information into “quickly and easily grasped health meaning, making healthier
options unmistakable” (109). However, Daily Value systems provide information only and
provide no translation. Traffic light systems translate nutrient information into “green”,
“amber”, and “red” colours and provide health meaning, however it is unclear if such an
approach would make healthier options “unmistakable”. For example, when comparing a
product with three green and one red traffic light to a product with two green and two yellow
traffic lights, the healthier option may not be evident. As an alternative, the Institute of Medicine
has suggested that the FOP system most likely to enable healthier food choices would be (109):
1. Simple (requiring no nutritional knowledge or sophistication on the part of the
consumer to understand).
2. Interpretive (providing guidance based on nutrition information without providing any
specific nutrition information).
3. Ordinal (using a scaled or ranked approach).
As such the Institute of Medicine has proposed a FOP system that uses 0-3 stars based on the
food’s content of nutrients of public health concern, but does not explicitly mention these
nutrients in the symbol or the amount of these nutrients a serving of the food provides. While
based on the best available evidence, the Institute of Medicine’s proposed FOP system has not
been tested with consumers.
109
7.2.2 Research
Although our research has pointed to several possible policy directions for FOP systems in
Canada, we have concluded that the preponderance of evidence supports the adoption of a single,
standardized, and mandatory FOP system with a common nutrient profiling scheme. However,
continued research is needed to support the development of policy in this area. Further, any
evolutions in FOP system policy will need to be followed up with evaluation. Our above
discussion leads to a number of valuable research questions.
From this research, it is clear that nutrient profiling criteria to define “healthy” to form the basis
of either a single, standardized FOP system or existing third-party and manufacturer developed
FOP systems is needed. Presently, there is no gold standard for defining “healthy” foods (95),
however as described in section 1.3.3, there are several important considerations in developing a
nutrient profiling system. Considerations include the nutrient profiling system’s purpose and
target audience, application, included nutrients and food components, amount of food, type of
model, and basis of the system (19,20,93-95,105,106). The Institute of Medicine has proposed
that nutrient profiling criteria to support FOP systems be developed on the basis of existing
nutrient-content and health claims (109). Regardless of whether criteria are developed based on
existing regulations as proposed, or are modeled after existing nutrient profiling systems
developed internationally, or are based on current healthy eating guidance, future investigations
should focus on the creation and validation of a nutrient-profiling criteria to support FOP
systems in Canada, or to support healthy eating by Canadian consumers.
Investigations into how proposed nutrient profiling criteria may impact the food supply and
nutrient intake of Canadians will also be needed. As in our Chapter 4 investigation, proposed
nutrient profiling criteria could be applied to the existing Canadian food supply to determine the
proportion of products already meeting the criteria based on their existing composition. This
would permit the evaluation of the potential of different nutrient profiling criteria to stimulate
reformulation. Further, investigations will be needed to determine if proposed nutrient profiling
criteria for Canada will support improved dietary intakes and health. The potential impact of
foods that meet third-party or industry summary indicator system nutrient-profiling criteria on
health risk factors and outcomes, and dietary intakes has been explored in a handful of studies,
with positive results (165-169). However, the potential impact of nutrient-specific systems –
110
such as traffic light nutrient profiling criteria or nutrient-content claim criteria – on dietary
intakes have not been studied.
Related to FOP system format, although our Chapter 6 investigation singled out nutrient-specific
systems as the preferred format these systems share many common features with the existing NFt
(153). Common features include complex nutrition terminology as well as high numeracy
requirements related to using serving size, percent daily value, and nutrient content information.
As described in section 1.2.1.1 these features have been an impediment to the use and
understanding the existing mandatory nutrition label and have even prompted some of the calls
to implement a single, standard FOP system. Research has shown that consumers with the
lowest levels of nutrition and health knowledge and infrequent label readers found nutrient
specific systems more challenging to understand than summary indicator systems (63), and these
consumers preferred summary indicator systems to nutrient specific systems (138). Mejean et al.
(138) found that, compared to consumers with higher levels of education, a greater proportion of
consumers with lower education levels preferred summary indicator systems over nutrient-
specific systems. The population surveyed in our Chapter 6 investigation had higher levels of
education as well as a demonstrated interest in nutrition and greater tech literacy through their
regular participation in online on food and nutrition surveys. As a result, our population’s
preference for nutrient-specific FOP systems may not be generalizable to population subgroups
with lower levels of education and nutrition knowledge – who may be in most need of help.
Investigations that explore preferences and understanding of FOP systems among consumers
with lower levels of education and health literacy will be needed prior to identifying an
appropriate standard FOP system for Canada. Focus group discussions may be a more
appropriate mean to get at this needs and concerns of these population subgroups. The nutrient-
specific systems evaluated in our study should also be tested against the simple, interpretive, and
ordinal FOP system proposed by the US Institute of Medicine, which was published after our
study was completed.
Research to date evaluating the effectiveness of FOP systems has been deemed to be of low
methodological quality (170). Self-reported, observational studies, and survey experiments, like
the one reported on in this thesis, need to be followed up with studies that examine end points
such as sales, reformulation, and health outcomes. However, the majority of research examining
these outcomes has been rated as low to mediocre (170). Natural experiments that include a
111
control are needed to build upon the existing FOP sales studies. Further, prospective studies of
reformulation in response to FOP systems are needed. Most existing studies have retrospectively
asked manufacturers participating in third-party FOP programs about reformulation in response
to involvement in the program (112-115). Ideally, prospective studies should include both
participating and non-participating manufacturers to determine if reformulation can be attributed
to participating in the FOP system. Moreover, a prospective study would provide evidence of
the impact of voluntary FOP systems on the entire food supply. Finally, Vyth et al. (170) stated
that the best evidence of FOP system effectiveness will be generated by a longitudinal,
randomized controlled trial in a real-life setting that examines the health effects of FOP systems
on biomarkers of health and nutritional status. While this would be the gold standard, the
feasibility of such an approach is questionable as it would be difficult both to establish an
appropriate control group and conduct a study of sufficient duration to demonstrate an effect and
attribution of FOP labelling to health outcomes would be difficult to quantify separately from
other risk factors.
112
Chapter 8
8 Summary and Conclusions
FOP systems are widespread in the Canadian marketplace, yet there are no specific regulations
governing their use. The Canadian Government has expressed its intentions to maintain the
status quo with respect to FOP systems in Canada. There is ample evidence regarding the
challenges of using the current NFt; however, data are lacking regarding FOP systems in the
Canadian context, and it is unclear if Canadians are being well served by the current minimal
policies regarding FOP systems in Canada. Further, Canadians’ liking and understanding of
different FOP systems have not been studied. Thus, the overall purpose of this research was to
generate food supply and consumer data to support the evolution of Canadian food labelling
policy as it relates to FOP systems.
We observed that although a large proportion of Canadian food products qualify for common
third-party and manufacturer developed summary indicator FOP systems, few products that meet
the systems criteria are identified by their symbols (Chapter 4). Given this, we concluded that
these symbols provide minimal impetus for reformulation for prospective participants. Further,
we concluded that consumers cannot rely on these summary indicator systems to identify
products of superior nutritional quality.
Additionally, in Chapter 5 we observed that products with FOP systems were similar in calorie
and nutrient composition to products without FOP systems. Further, we found that no specific
FOP system type was a better indicator of products of higher nutritional quality. Given the
similar nutrient profiles of foods with and without FOP symbols, we concluded such symbols
were deceptive and misleading consumers. Therefore, our results suggest that the introduction of
minimum nutrient-profiling criteria for all FOP systems may be an appropriate policy option to
minimize this problem.
In our third study (Chapter 6) we demonstrated that consumers perceived a frozen meal and
breakfast cereal as healthier and/or lower in negative nutrients and higher in positive nutrients
when they carried FOP systems. For the most part, this effect was attenuated in the presence of
the NFt – supporting the concerns that FOP systems have the potential to mislead when used
113
without the mandatory nutrition label, particularly when they truncate the search for additional
information found in the NFt. Our study was the first to test Canadians liking and understanding
of different FOP systems. These novel data suggest that nutrient-specific FOP systems should be
explored as possible single, standardized FOP labelling system for Canadians.
When our results are considered in relation to other studies of FOP systems, they suggest that
Canadians may not be well served by Canada’s existing policies on FOP systems. FOP system
policy in Canada needs to evolve to include standard minimum nutrient profile criteria, and
perhaps the implementation of a single, standardized nutrient-specific system for use on food
labels.
114
References
(1) World Health Organization. Diet, nutrition and the prevention of chronic diseases. World Health
Organization technical report series 2003;916.
(2) The Secretariat for the Intersectoral Healthy Living Network in partnership with the F/P/T Healthy
living Task Group and the F/P/T Advisory Committee on Population Health and Health Security. The
Integrated Pan-Canadian Healthy Living Strategy. 2005.
(3) World Health Organization. Global Strategy on Diet, Physical Activity and Health. 2004.
(4) Health Canada. Eating Well with Canada's Food Guide. 2007.
(5) Government of Canada. Regulations Amending the Food and Drug Regulations (Nutrition Labelling,
Nutrient Content Claims and Health Claims). The Canada Gazette, Part II 2003;137(1):154.
(6) Canadian Council of Food and Nutrition. Tracking Nutrition Trends VII. 2008.
(7) Ni Mhurchu C, Gorton D. Nutrition labels and health claims in New Zealand and Australia; a review
of use and understanding. Aust NZ J Public Health 2007;31:105-112.
(8) Gorton D. Nutrition labelling - Update of scientific evidence on consumer use and understanding of
nutrition labels and claims. 2007.
(9) Cowburn G, Stockley L. Consumer understanding and use of nutrition labelling: A systematic review.
Public Health Nutr 2005;8(1):21-28.
(10) van Kleef E, van Trijp H, Paeps F, Fernández-Celemín L. Consumer preferences for front-of-pack
calories labelling. Public Health Nutr 2007;11(2):203-213.
(11) The Strategic Counsel. Focus Testing of Creatives for the Nutrition Facts Education Initiative.
2010;HC POR 09-16.
(12) Emrich TE, L'Abbe MR. Nutrition Labelling Education Research Synthesis. 2014 Janaury 2014.
(13) The Standing Committee on Health. Healthy Weights for Healthy Kids. 2007.
(14) Sodium Working Group. Sodium Reduction Strategy for Canada: Recommendations of the Sodium
Working Group. 2010.
115
(15) European Food Information Council. Global Update on Nutrition Labelling. 2013 February 2013.
(16) Department of Health, Food Standards Agency, Welsh Government, The Scottish Government.
Guide to creating a front of pack (FoP) nutrition label for pre-packed products sold through retail outlets.
2013.
(17) Plibersek T, Neumann S. Star Rating System for Packaged Foods. Joint Media Release 2013 2013
Jun 14.
(18) Health Canada. Managing health claims for foods in Canada: Towards a modernized framework.
2007.
(19) Committee on the Examination of Front-of-Package Nutrition Rating Systems and Symbols, Institute
of Medicine. Examination of Front-of-Package Nutrition Rating Systems and Symbols: Phase I Report.
2010.
(20) Lobstein T, Davies S. Defining and labelling ‘healthy’ and ‘unhealthy’ food. Public Health Nutr
2009;12(3):331-340.
(21) Carlson LA. Front-of-the-Pack and On-Shelf Labeling: Tools for Spotting Nutritious Choices at the
Supermarket Shelf. Nutr Today 2010;45(1):15-21.
(22) Lupton JR, Balentine DA, Black RM, Hildwine R, Ivens BJ, Kennedy ET, et al. The Smart Choices
front-of-package nutrition labeling program: rationale and development of the nutrition criteria. Am J
Clin Nutr 2010;91(4):1078S-1089S.
(23) World Health Organization. Global status report on non-communicable diseases 2010. 2011;
Available at: Global status report on non-communicable diseases 2010. Accessed January 29, 2014.
(24) Mirolla M. The Cost of Chronic Disease in Canada. 2004.
(25) Beaglehole R, Bonita R, Horton R, Adams C, Alleyne G, Asaria P, et al. Priority actions for the non-
communicable disease crisis. Lancet 2011;377:1438-1447.
(26) Canadian Food Inspection Agency. Guide to Food Labelling and Advertising. 2010; Available at:
http://www.inspection.gc.ca/english/fssa/labeti/guide/ch1e.shtml. Accessed October 27, 2010.
(27) Codex Alimentarius Commission. Guidelines on Nutrition Labelling. 2012; Available at:
www.codexalimentarius.net/download/standards/34/CXG_002e.pdf. Accessed January 29, 2014.
116
(28) Environics Research Group. Post-Campaign Evaluation – Healthy Eating. 2008.
(29) The Strategic Counsel. Canadians' Understanding and Use of the Nutrition Facts table: Baseline
National Survey Results. 2011;POR 031-10(HCPOR-10-06).
(30) The Strategic Counsel. Canadians' Understanding and Use of the Nutrition Facts table: Survey
Results. 2012;POR-088-11(HCPOR-11-01).
(31) US Department of Health and Human Services, Centers for Disease Control and Prevention. Healthy
People 2000 Final Review. 2001.
(32) Drichoutis AC, Lazaridis P, Nayga RM. Who is looking for nutritional food labels? Eurochoices
2005;4:18-23.
(33) Gorton D, Ni Mhurchu C, Chen M, Dixon R. Nutrition labels: a survey of use, understanding and
preferences among ethnically diverse shoppers in New Zealand. Public Health Nutr 2008;12(9):1359-
1365.
(34) Miller C, Brown J. Knowledge and use of the food label among senior women in the management of
type 2 diabetes mellitus. Journal of Nutrition, Health & Aging 1999;3(3):152-7.
(35) Rothman RL, Housam R, Weiss H, Davis D, Gregory R, Gebretsadik T, et al. Patient understanding
of food labels: The role of literacy and numeracy. American Journal of Preventive Medicine
2006;31(5):391-398.
(36) Jay M, Adams J, Herring SJ, Gillespie C, Ark T, Feldman H, et al. A randomized trial of a brief
multimedia intervention to improve comprehension of food labels. Prev Med 2009 200901;48(1):25-31.
(37) Kristal AR, Levy L, Patterson RE, Li SS, White E. Trends in food label use associated with new
nutrition labeling regulations. American Journal of Public Health 1998;88(8):1212-1215.
(38) Pelletier AL, Chang WW, Delzell JEJ, McCall JW. Patients' Understanding and Use of Snack Food
Package Nutrition Labels. The Journal of the American Board of Family Practice 2004;17(5):319-323.
(39) Hawthorne KM, Moreland K, Griffin IJ, Abrams SA. An Educational Program Enhances Food Label
Understanding of Young Adolescents. J Am Diet Assoc 2006;106(6):913-916.
(40) Lewis CJ, Yetley Ea. Nutrition labels in bar graph format deemed most useful for consumer purchase
decisions using adaptive conjoint analysis. Journal of the American Dietetic Association 1992;92(1):62.
117
(41) Hrovat KB, Harris KZ, Leach AD, Russell BS, Harris BV, Sprecher DL. The new food label, type of
fat, and consumer choice. A pilot study. Archives of family medicine 1994;3(8):690-695.
(42) Dooley DA, Novotny R, Britten P. Integrating research into the undergraduate nutrition curriculum:
Improving shoppers’ awareness and understanding of nutrition facts labels. J Nutr Educ 1998;30:225-231.
(43) Levy AS, Fein SB. Consumers' Ability to Perform Tasks Using Nutrition Labels. J Nutr Educ
1998;30(4):210-217.
(44) Miller CK, Jensen GL, Achterberg CL. Evaluation of a Food Label Nutrition Intervention for
Women with Type 2 Diabetes Mellitus. J Am Diet Assoc 1999;99(3):323-328.
(45) Levy L, Patterson RE, Kristal AR, Li SS. How well do consumers understand percentage daily value
on food labels? American Journal of Health Promotion 2000;14(3):157-160.
(46) Sullivan AD. Determining how low-income food shoppers perceive, understand, and use food labels.
Canadian Journal of Dietetic Practice and Research 2003;64(1):25-30.
(47) Misra R. Knowledge, Attitudes, and Label Use among College Students. J Am Diet Assoc
2007;107(12):2130-2134.
(48) NRG Research Group. Qualitative Study on Use and Understanding of Nutrition Labelling. 2007.
(49) Ascentum. Nutrition Labelling Resources/Tools Consultation. 2009.
(50) Phase 5. Nutrition Facts Education Campaign Web site Usability Testing. 2010;HCPOR-10-05.
(51) Lubman N, Doak C, Jasti S. Food Label Use and Food Label Skills among Immigrants from the
Former Soviet Union. Journal of Nutrition Education and Behavior 2012 201209/10;44(5):398-406.
(52) Vanderlee L, Goodman S, Yang WS, Hammond D. Consumer understanding of calorie amounts and
serving size: implications for nutritional labelling. Canadian Journal of Public Health 2012 September;
2013/11;103:327+.
(53) Lando AM, Lo SC. Single-Larger-Portion-Size and Dual-Column Nutrition Labeling May Help
Consumers Make More Healthful Food Choices. Journal of the Academy of Nutrition and Dietetics 2013
201302;113(2):241-250.
118
(54) Lando AM, Labiner-Wolfe J. Helping consumers make more healthful food choices: Consumer
views on modifying food labels and providing point-of-purchase nutrition information at quick service
restaurants. Journal of Nutrition Education and Behaviour 2007;39(3):157-163.
(55) Levy AS, Fein SB, Schucker RE. More effective nutrition label formats are not necessarily preferred.
Journal of the American Dietetic Association 1992;92(10):1230-1234.
(56) Silverglade B, Ringel Heller I. Food Labelling Chaos - The Case for Reform. 2010.
(57) Turnbull J. Canadian Medical Associations's Presentation to the House of Commons Standing
Committee on Health Nutrition Labelling. 2011 March 3.
(58) Jeffery B. Letter to Federal Minister of Health: Ensure that the Food Label Modernization initiative
includes health and nutrition issues to protect families, save health care resources, and raise workforce
productivity by disease prevention. Centre for Science in the Public Interest 2013 October 16.
(59) Burton S, Andrews JC. Age, product nutrition, and label format effects on consumer perceptions and
product evaluations. The Journal of Consumer Affairs 1996 Summer 1996;30(1):68.
(60) Jones G, Richardson M. An objective examination of consumer perception of nutrition information
based on healthiness ratings and eye movements. Public Health Nutr 2007 Mar.;10(3):238-244.
(61) Miller C, Probart C, Achterberg C. Knowledge and Misconceptions About the Food Label Among
Women With Non-Insulin-Dependent Diabetes Mellitus. Diabetes Educ 1997;23(4):425-432.
(62) Signal L, Lanumata T, Robinson J, Tavila A, Wilton J, Ni Mhurchu C. Perceptions of New Zealand
nutrition labels by Maori, Pacific and low-income shoppers. Public Health Nutrition 2007;11(7):706-713.
(63) Feunekes GIJ, Gortemaker IA, Willems AA, Lion R, van den Kommer M. Front-of-pack nutrition
labelling: Testing effectiveness of different nutrition labelling formats front-of-pack in four European
countries. Appetite 2008;50(1):57-70.
(64) Borgmeier I, Westenhoefer J. Impact of different food label formats on healthiness evaluation and
food choice of consumers: A randomized-controlled study. BMC Public Health 2009;9:184.
(65) Kelly B, Hughes C, Chapman K, Louie JC, Dixon H, Crawford J, et al. Consumer testing of the
acceptability and effectiveness of front-of-pack food labelling systems for the Australian grocery market.
Health Promot Int 2009;24(2):120-129.
119
(66) McCullum C, Achterberg CL. Food shopping and label use behavior among high school-aged
adolescents. Adolescence 1997;32(125):181-197.
(67) Wansink B. How Do Front and Back Package Labels Influence Beliefs About Health Claims? The
Journal of Consumer Affairs 2003;37(2):305-316.
(68) Grunert KG, Wills JM. A review of European research on consumer response to nutrition
information on food labels. J Public Health 2007;15(5):385-399.
(69) Colby SE. Nutrition Marketing on Food Labels. Proceedings of the Academy of Marketing Studies
2008;13(1):17-21.
(70) Colby SE, Johnson L, Hoverson B. Nutrition marketing on food labels. J Nutr Educ Behav
2010;42(2):92-98.
(71) Health Canada. Substantiation of Health Claims (Scientific Evidence). 2014; Available at:
http://www.hc-sc.gc.ca/fn-an/label-etiquet/claims-reclam/assess-evalu/index-eng.php. Accessed Feb 12,
2014.
(72) Health Canada. Summary of Health Canada's Assessment of a Health Claim about Ground Whole
Flaxseed and Blood Cholesterol Lowering. 2014; Available at: http://www.hc-sc.gc.ca/fn-an/label-
etiquet/claims-reclam/assess-evalu/flaxseed-graines-de-lin-eng.php. Accessed Feb 12, 2014.
(73) Government of Canada. Food and Drugs Act. 2010.
(74) Health Canada. Guidance Document – The Use of Probiotic Microorganisms in Food 2009 April
2009.
(75) LeGault L, Brandt MB, McCabe N, Adler C, Brown A, Brecher S. 2000-2001 Food Label and
Package Survey; An Update on the Prevalence of Nutrition Labeling and Claims on Processed, Packaged
Foods. Journal of the American Dietetic Association 2004;104(6).
(76) Brandt M, Moss J, Ferguson M. The 2006-2007 food label and package survey (FLAPS): nutrition
labeling, trans fat labeling. Journal of Food Composition and Analysis 2009 12;22:S74-S77.
(77) Brandt MB, Moss J, Ellwood K, Ferguson M, Asefa A. Tracking Label Claims. Chicago 2010
20100101;64(1):34.
120
(78) Schermel A, Emrich T, Arcand J, Wong C, L’Abbe M. Nutrition Marketing on Processed Food
Packages in Canada: 2010 Food Label Information Program. Appl Physiol Nutr Metab 2013;38(6):666-
672.
(79) Williams P, Yeatman H, Ridges L, Houston A, Rafferty J, Roasler A, et al. Nutrition function, health
and related claims on packaged Australian food products - prevalence and compliance with regulations.
Aia Pac J Clin Nutr 2006;15(1):10-20.
(80) Bureau Europeen des Unions de Consommateurs. Report on European Consumers’ Perception of
Foodstuffs Labelling. 2005.
(81) Canadian Council of Food and Nutrition. Ethnographic Study: A New Perspective on Canadians'
Attitudes and Behaviours Toward Food and Nutrition. 2010.
(82) Andrews JC, Netemeyer R, Burton S. Consumer generalization of nutrient content claims in
advertising. Journal of Marketing 1998;62:62-75.
(83) Roe B, Levy AS, Derby BM. The impact of health claims on consumer search and product
evaluations outcomes: Results from FDA experimental data. Journal of Public Policy and Marketing
1999;18(1):89-105.
(84) Kozup JC, Creyer EH, Burton S. Making Healthful Food Choices: The Influence of Health Claims
and Nutrition Information on Consumers' Evaluations of Packaged Food Products and Restaurant Menu
Items. Journal of Marketing 2003;67(2):19-34.
(85) van trijp, Hans C. M., van der Lans, Ivo A. Consumer perceptions of nutrition and health claims.
Appetite 2007;48:305-324.
(86) Labiner-Wolfe J, Lin CJ, Verrill L. Effect of Low-carbohydrate Claims on Consumer Perceptions
about Food Products’ Healthfulness and Helpfulness for Weight Management. Journal of Nutrition
Education and Behavior 2010;42(5):315-320.
(87) Lähteenmäki L, Lampila P, Grunert K, Boztug Y, Ueland Ø, Åström A, et al. Impact of health-
related claims on the perception of other product attributes. Food Policy 2010;35(230):239.
(88) Zank GM, Kemp E. Examining Consumers’ Perceptions of the Health Benefits of Products with
Fiber Claims. The Journal of Consumer Affairs 2012;46(2):333-344.
121
(89) Wong CL, Arcand J, Mendoza J, Henson SJ, Qi Y, Lou W, et al. Consumer attitudes and
understanding of low sodium claims on food: An analysis of healthy and hypertensive individuals.
American Journal of Clinical Nutrition 2013;97(6):1288-1298.
(90) Government of Canada. Food and Drug Regulations. 2010.
(91) Verbekea W, Scholdererb J, Lahteenmakib LL. Consumer appeal of nutrition and health claims in
three existing product concepts. Appetite 2009;52:684-692.
(92) Dietitians of Canada. Diabetes, Obesity and Cardiovascular Disease Network. Evidence-Based
Background Paper on Point-of-Purchase Nutrition Programs. 2006.
(93) Stockley L. Review of 'front of pack' nutrition schemes. 2007.
(94) Reza Z. Defining “Healthy” Foods Environmental Scan of the Situation in Canada. 2009.
(95) World Health Organization. Nutrient Profiling: Report of a WHO/IASO Techinical Meeting. 2010
Oct 4-6.
(96) The 2007/08 Food Regulation Standing Committee, Front-of-Pack Labelling Working Group.
Information Prepared by The 2007/08 Food Regulation Standing Committee, Front-of-Pack Labelling
Working Group.
(97) Rayner M, Boaz A, Higginson C. Consumer use of health-related endorsements on food labels in the
United Kingdom and Australia. J Nutr Educ 2001;33(1):24-30.
(98) Vyth EL, Steenhuis IHM, Vlot JA, Wulp A, Hogenes MG, Looije DH, et al. Actual use of a front-of-
pack nutrition logo in the supermarket: consumers' motives in food choice. Public Health Nutr
2010;13(11):1882-1889.
(99) Williams P, Duncan R, de Agnoli K, Hull A, Owers A, Wang T. Introduction and use of front of
pack daily intake labelling on Australian packaged foods 2007-2009. Food Australia 2010;62(12):583-
588.
(100) Storcksdieck genannt Bonsmann S, Fernandez-Celemın L, Larranaga A, Egger S, Wills JM,
Hodgkins C, et al. Penetration of nutrition information on food labels across the EU-27 plus Turkey.
European Journal of Clinical Nutrition 2010;64:1379-1385.
122
(101) The Grocery Manufacturers Association, The Food Marketing Institute. Facts Up Front. Available
at: http://www.factsupfront.org. Accessed Feb 14, 2014.
(102) Heart and Stroke Foundation. Heart and Stroke Health Check Program. 2010; Available at:
http://www.healthcheck.org/page/what-health-check. Accessed July 12, 2011.
(103) NuVal LLC. NuVal Nutrition Made Easy. 2014; Available at: http://www.nuval.com/. Accessed
Feb 14, 2014.
(104) Whole Grains Council. Whole Grain Stamp. 2011; Available at:
http://www.wholegrainscouncil.org/whole-grain-stamp. Accessed July 13, 2011.
(105) Garsetti M, de Vries J, Smith M, Amosse A, Rolf-Pedersen N. Nutrient profiling schemes:
overview and comparative analysis<br />. European Journal of Clinical Nutrition 2007;46:15-28.
(106) Hawkes C. Defining “Healthy” and “Unhealthy”Foods: An International Review. 2009.
(107) Roodenburg AJC, Popkin BM, Seidell JC. Development of international criteria for a front of
package food labelling system: the International Choices Programme. European Journal of Clinical
Nutrition 2011;65:1190-1200.
(108) Trichterborn J, Harzer G, Kunz C. Nutrient profiling and food label claims: evaluation of dairy
products in three major European coutries. Eur J Clin Nutr 2011;65:1032-1038.
(109) Committee on Examination of Front-of-Package Nutrition Ratings Systems and Symbols (Phase II),
Institute of Medicine. Front-of-Package Nutrition Rating Systems and Symbols: Promoting Healthier
Choices. 2011 October 20.
(110) Brownell KD, Koplan JP. Front-of-Package Nutrition Labeling - An abuse of trust by the food
industry? N Engl J Med 2011;364(25):2373-2375.
(111) Health Canada. Summary of the Development and Use of a Surveillance Tool: The Classification of
Foods in the Canadian Nutrient File According to Eating Well with Canada's Food Guide. 2014;130517.
(112) Young L, Swinburn B. Impact of the Pick the Tick food information programme on the salt content
of food in New Zealand. Health Promotion International 2002;17(1):13-19.
123
(113) Williams P, McMahon A, Boustead R. A case study of sodium reduction in breakfast cereals and
the impact of the Pick the Tick food information program in Australia. Health Promotion International
2003;18(1):51-56.
(114) Vyth EL, Steenhuis IHM, Roodenburg AJC, Brug J, Seidell JC. Front-of-pack nutrition label
stimulates healthier product development: a quantitative analysis. Int J Behav Nutr Phys Act 2010;7.
(115) Dummer J. Sodium reduction in Canadian food products with the health check program. Can J Diet
Pract Res 2012;73(1):e227-e332.
(116) Louie JC, Dunford EK, Walker KZ, Gill TP. Nutritional Quality of Australian Breakfast Cereals.
Are They Improving? Appetite 2012;59:464-470.
(117) Hodgkins C, Barnett J, Wasowicz-Kirylo G, Stysko-Kunkowska M, Gulcan Y, Kustepeli Y, et al.
Understanding how consumers categorise nutritional labels: A consumer derived typology for front-of-
pack labelling. Appetite 2012;59:806-817.
(118) Sutherland LA, Kaley LA, Fischer L. Guiding Stars: the effect of a nutrition navigation program on
consumer purchases at the supermarket. Am J Clin Nutr 2010 20100401;91(4):1090S.
(119) Hawkes C. Nutrition labels and health claims: the global regulatory environment. 2004.
(120) Scott V, Worsley AF. Ticks, claims, tables and food groups: A comparison for nutrition labelling.
Health Promot Int 1994;9(1):27-37.
(121) Reid RD, Slovinec D'Angelo ME, Dombrow CA, Heshka JT, Dean TR. The Heart and Stroke
Foundation of Canada's Health Check food information program: modelling program effects on consumer
behaviour and dietary practices. Can J Public Health 2004 Mar-Apr;95(2):146-150.
(122) Steenhuis IHM, Kroeze W, Vyth EL, Valk S, Verbauwen R, Seidell JC. The effects of using a
nutrition logo on consumption and product evaluation of a sweet pastry. Appetite 2010;55(3):707-709.
(123) Andrews JC, Burton S, Kees J. Is simpler always Better? Consumer evaluations of front-of-package
nutrition symbols. JPP&M 2011.
(124) Newman CL, Howlett E, Burton S. Shopper Response to Front-of-Package Labeling Programs:
Potential Consumer and Retail Store Benefits. Journal of Retailing 2014;90(1):13-26.
124
(125) Roberto CA, Shivaram M, Martinez O, Boles C, Harris JL, Brownell KD. The Smart Choices front-
of-package nutrition label. Influence on perceptions and intake of cereal. Appetite 2012;58:651-657.
(126) Hawley KL, Roberto CA, Bragg MA, Liu PJ, Schwatz MB, Brownell KD. The science of front-of-
package labels. Public Health Nutr 2012.
(127) Temple JL, Johnson KM, Archer K, LaCarte A, Yi C, Epstein LH. Influence of simplified nutrition
labeling and taxation on laboratory energy intake in adults. Appetite 2011;57:184-192.
(128) Sacks G, Rayner M, Swinburn B. Impact of front-of-pack ‘traffic-light’ nutrition labelling on
consumer food purchases in the UK. Health Promot Int 2009 2009;24(4):344-352.
(129) Sacks G, Tikellis K, Miller L, Swinburn B. Impact of 'traffic-light' nutrition information on online
food purchases in Australia. Aust NZ J Public Health 2011;35(2):122-126.
(130) Neuhouser ML, Kristal AR, Patterson RE. Use Of Food Nutrition Labels is Associated with Lower
Fat Intake. J Am Diet Assoc 1999;99(1):45-53.
(131) Draper AK, Adamson AJ, Clegg S, Malam S, Rigg M, Duncan S. Front-of-pack nutrition labelling:
are multiple formats a problem for consumers. European Journal of Public Health 2011;23(3):517-521.
(132) Malam S, Clegg S, Kirwan S, McGinigal S. Comprehension and use of UK nutrition signpost
labelling schemes. 2009.
(133) Schmidt S. Aglukkaq pans U.S. nutrition recommendations. Postmedia News 2011 October 20.
(134) Grunert KG, Fernández-Celemín L, Wills JM, Bonsmann SSg, Nureeva L. Use and understanding
of nutrition information on food labels in six European countries. J Public Health 2010 Jun;18(3):261-
277.
(135) Grunert KG, Wills JM, Fernández-Celemín L. Nutrition knowledge, and use and understanding of
nutrition information on food labels among consumers in the UK. Appetite 2010 Oct;55(2):177-189.
(136) Berning JP, Chouinard HH, McCluskey JJ. Consumer Preferences for Detailed versus Summary
Formats of Nutrition Information on Grocery Store Shelf Labels<br />. Journal of Agricultural & Food
Industrial Organization 2008;6(1):1542.
(137) Moser A, Hoefkens C, Van Camp J, Verbeke W. Simplified nutrient labelling: Consumers'
perceptions in Germany and Belgium. J Consum Prot Food Saf 2010 May;5(2):169-180.
125
(138) Mejean C, Macouillard P, Peneau S, Hercberg S, Castetbon K. Perceptions of front-of-pack labels
according to social characteristics, nutritional knowledge and food purchasing habits. Public Health Nutr
2012;16(3):392-402.
(139) Maubach N, Hoek J. A qualitative study of New Zealand parents' views on front-of-pack nutrition
labels. Nutrition & Dietetics 2010;67:90-96.
(140) Carter O, Mills B, Phan T. An independent assessment of the Australian food industry's Daily
Intake Guide 'energy alone' label. Health Promotion Journal of Australia 2011;22(1):63-67.
(141) Larsson I, Lissner L, Wilhelmsen L. The 'Green Keyhole' revisited: Nutritional knowledge may
influence food selection. European Journal of Clinical Nutrition 1999;52:777-780.
(142) Evidence for Health Claims on Food: How Much is Enough? Proceeding of the Symposium
Presented at the Canadian Nutrition Congress. ; June 21, 2007; ; 2008.
(143) Food Manufacturers - Heart&Stroke Health Check™ Program. 2010; Available at:
http://www.healthcheck.org/page/licensee-overview-0.
(144) Kraft Canada Inc. Sensible Solution. 2011; Available at: http://www.kraftcanada.com/en/healthy-
living-ideas/sensiblesolution/sensiblesolution.aspx. Accessed July 12, 2011.
(145) Heart and Stroke Foundation. Health Check™ Nutrient Criteria: December 2009. 2009.
(146) Old Dutch Foods I. Snack Wise. 2011; Available at: http://www.olddutchfoods.com/nutrition-
highlights/snack-wise. Accessed October 3, 2011.
(147) Viera AJ, Garrett JM. Understanding Interobserver Agreement: The Kappa Statistic. Family
Medicine 2005;37(5):360-363.
(148) Heart and Stroke Foundation. Health Check™ Nutrient Criteria: July 2009. 2009.
(149) Heart and Stroke Foundation. Health Check Nutrient Criteria: January 2011. 2011.
(150) Pantazaopoulos P, Kwong K, Lillycrop W, Gao Y, Samadhin M, Ratnayake WM, et al. Trans and
saturated fat on food labels in Canada: fact or fiction? Can J Public Health 2011;102(4):313-316.
(151) World Health Organization. Preventing chronic diseases: a vital investment. 2005.
(152) Mintel Global Market Navigator. Grocery Retailing - Canada - a snapshot. 2010.
126
(153) Emrich TE, Qi Y, Mendoza JE, Lou W, Cohen JE, L'Abbe MR. Consumer perceptions of the
Nutrition Facts table and Front-of-Pack nutrition rating systems. Appl Physiol Nutr Metab 2014;39:1-8.
(154) Arcand J, Mendoza J, Qi Y, Henson S, Lou W, L'Abbe MR. Results of national survey examining
Canadians' concern, actions, barriers, and support for dietary sodium reduction interventions. Can J
Cardiol 2013 May;29(5):628-631.
(155) Emrich TE, Mendoza JE, L'Abbe MR. Effectiveness of Front-of-pack nutrition symbols: A pilot
study with consumers. Can J Diet Pract Res 2012;73(4).
(156) Blitstein JL, Evans WD. Use of Nutrition Facts Panels among Adults Who Make Household Food
Purchasing Decisions. Journal of Nutrition Education and Behavior 2006;38(6):360-364.
(157) Campos S, Doxey J, Hammond D. Nutrition labels on pre-packaged foods: a systematic review.
Public Health Nutr 2011 August;14:1506.
(158) Emrich TE, Cohen JE, Lou WY, L'Abbe MR. Food products qualifying for versus carrying a front-
of-pack nutrition rating system: a cross-sectional study. BMC Public Health 2013;13(846).
(159) Health Canada. Consulting Canadians to Modermize and Improve Food Labels. 2014; Available at:
http://www.hc-sc.gc.ca/fn-an/alt_formats/pdf/label-etiquet/modernize-report-moderniser-rapport-eng.pdf.
Accessed July 12, 2014.
(160) Savoie N, Barlow K, Harvey KL, Binnie MA, Pasut L. Consumer Perceptions of Front-of-packahe
Labelling Systems and Healthiness of Foods. Canadian Journal of Public Health 2013;104(5):e359-e363.
(161) Heart and Stroke Foundation. Backgrounder - Health Check. 2010.
(162) Health Canada. Nutrition Labelling Consultation. 2014; Available at: http://www.hc-sc.gc.ca/fn-
an/label-etiquet/consultation/index-eng.php. Accessed July 16, 2014.
(163) The Trans Fat Task Force. TRANSforming the Food Supply: Report of the Trans Fat Task Force
Submitted to the Minister of Health. 2006.
(164) von Tigerstrom B. Food Labelling for Healthier Eating: Is Front-of-package Labelling the Answer?
Manitoba Law Journal 2009;33(1).
(165) Roodenburg AJC, Temme EHM, Davies OH, Seidell JC. Potential impact of the Choices
Programme on nutrient intakes in the Dutch population. Nutr Bull 2009 September 2009;34(3):318-323.
127
(166) Temme EHM, van der Voet H, Roodenburg AJC, Bulder A, van Donkersgoed G, van Klaveren J.
Impact of foods with health logo on saturated fat, sodium and sugar intake of young dutch adults. Public
Health Nutr 2010;14(4):635-644.
(167) Chiuve SE, Sampson L, Willet WC. Adherence to the Overall Nutritional Quality Index and Risk of
Total Chronic Disease. Am J Prev Med 2011 May;40(5):505-513.
(168) Vyth EL, Hendriksen MAH, Roodenburg AJC, Steenhuis IHM, van Raaij JMA, Verhagen H, et al.
Consuming a diet complying with front-of-pack label criteria may reduce cholesterol levels: a modeling
study. Eur J Clin Nutr 2012;66:510-516.
(169) Lichtenstein AH, Carson JS, Johnson RK, Kris-Etherton PM, Pappas A, Rupp L, et al. Food-intake
patterns assessed by using front-of-pack labeling program criteria associated with better diet quality and
lower cardiometabolic risk. Am J Clin Nutr 2014;99:454-462.
(170) Vyth EL, Steenhuis IHM, Brandt HE, Roodenburg AJC, Brug J, Seidell JC. Methodological quality
of front-of-pack labeling studies: a review plus identification of research challenges. Nutrition Reviews
2012;70(12):709-720.
128
Appendices
129
Appendix A
Proportion of food products that qualified for Heal th Check™ compared to the proportion of food produc ts that
carried the system's symbol by subcategory (N=7503)
Category N Products carrying Health Check™ N (%)
Products qualifying for Health Check™ N (%)
P-Value* Kappa Statistic (confidence interval)†
Vegetables & Fruit
Fruit Juices
Fresh Fruit
Frozen Fruit
Canned Fruit
Apple and other fruit sauces
Dried Fruit Pieces
Dried Fruit Snacks
Fresh and Frozen Vegetables (plain)
Canned Vegetables (plain)
Canned Tomatoes
Frozen and Canned Vegetables (seasoned, sauced)
Tomato Juice
Vegetable Juices and Blends
Tomato Paste
Minor Main Entrée Vegetable-Based Sauces
554
0
6
129
33
69
38
171
162
60
27
6
35
11
93
47
-
0
19
6
7
14
25
7
12
1
1
6
0
1
(8.5)
(-)
(0.0)
(14.7)
(18.2)
(10.1)
(36.8)
(14.6)
(4.3)
(20.0)
(3.7)
(16.7)
(17.1)
(0.0)
(1.1)
317
-
6
115
18
68
14
171
160
58
6
5
18
10
49
(57.2)
(-)
(100.0)
(89.2)
(54.6)
(98.6)
(36.8)
(100.0)
(98.8)
(96.7)
(22.2)
(83.3)
(51.4)
(90.9)
(52.7)
<.0001
-
0.0313
<.0001
0.0005
<.0001
.
<.0001
<.0001
<.0001
0.0625
0.1250
0.0005
0.0020
<.0001
0.1(0.1-0.2)
-
0.0 (0.0-0.0)
0.0 (0.0-0.1)
0.3(0.1-0.5)
0.0 (-0.0-0.0)
1.0 (1.0-1.0)
0.0 (0.0-0.0)
0.0 (-0.0-0.0)
0.0 (-0.0-0.0)
0.2 (-0.2-0.6)
0.1 (-0.1-0.3)
0.3 (0.1-0.6)
-0.0 (-0.0--0.0)
0.0 (-0.0-0.1)
130
Category N Products carrying Health Check™ N (%)
Products qualifying for Health Check™ N (%)
P-Value* Kappa Statistic (confidence interval)†
Vegetable-Based Dips
Frozen Fruit Bars
23
7
0
0
(0.0)
(0.0)
6
0
(26.1)
(0.0)
0.0313
-
0.0 (0.0-0.0)
-
Grain Products
Bread
Bread Products
Hot Breakfast Cereals
Breakfast Cereals (20-42 g per 250 mL)
Breakfast Cereals (≥43 g per 250 mL)
Very High Fibre Breakfast Cereals
Flour / Grains
Crackers / Rusks
Croutons
Rice Cakes
Waffles / Pancakes
Rice / Grains (plain)
Instant Rice (plain)
Pasta
Side Dishes – Rice, grains or potatoes (seasoned, sauced)
Side Dishes – Pasta or noodles (seasoned, sauced)
Grain-based Bars
185
228
57
79
145
7
6
261
53
62
73
78
5
383
132
104
178
20
17
4
7
7
2
0
28
5
2
2
8
2
32
7
0
7
(10.8)
(7.5)
(7.0)
(8.9)
(4.8)
(28.6)
(0.0)
(10.7)
(9.4)
(3.2)
(2.7)
(10.3)
(40.0)
(8.4)
(5.3)
(0.0)
(3.9)
112
91
40
29
53
7
3
115
6
18
9
78
5
242
31
1
52
(60.5)
(39.9)
(70.2)
(36.7)
(36.6)
(100.0)
(50.0)
(44.1)
(11.3)
(29.0)
(12.3)
(100.0)
(100.0)
(63.2)
(23.5)
(1.0)
(29.2)
<.0001
<.0001
<.0001
<.0001
<.0001
0.0625
0.2500
<.0001
1.0000
<.0001
0.0156
<.0001
0.2500
<.0001
<.0001
1.0000
<.0001
0.1 (0.1-0.2)
0.2 (0.1-0.3)
0.0 (-0.1-0.1)
0.3 (0.1-0.5)
0.1 (0.0-0.2)
0.0 (0.0-0.0)
0.0 (0.0-0.0)
0.2 (0.1-0.3)
0.9 (0.7-1.0)
0.2 (-0.0-0.3)
0.3 (-0.0-0.7)
0.0 (0.0-0.0)
0.0 (-0.0--0.0)
0.1 (0.1-0.1)
0.3 (0.1-0.4)
-0.0 (-0.0--0.0)
0.2 (0.1-0.3)
131
Category N Products carrying Health Check™ N (%)
Products qualifying for Health Check™ N (%)
P-Value* Kappa Statistic (confidence interval)†
Muffins / Muffin-Style Bars
Plain Popcorn
57
53
0
3
(0.0)
(5.7)
11
3
(15.8)
(5.7)
0.0010
1.0000
0.0 (0.0-0.0)
0.3 (-0.2-0.8)
Milk & Alternatives
Milk and Milk Based Drinks
Yogurts
Yogurt Based Drinks
Dairy-based dips
Fresh Cheese (plain and flavoured)
Cheese
Soy-based Cheese
Ricotta Cheese (plain)
Cottage Cheese (plain and flavoured)
Plant-based Beverages
82
95
17
48
63
390
5
10
23
76
1
12
0
0
0
9
0
0
0
8
(1.2)
(12.6)
(0.0)
(0.0)
(0.0)
(2.3)
(0.0)
(0.0)
(0.0)
(10.5)
57
49
11
37
0
46
0
1
18
61
(69.5)
(51.6)
(64.7)
(77.1)
(0.0)
(11.8)
(0.0)
(10.0)
(78.3)
(80.3)
<.0001
<.0001
0.0010
<.0001
-
<.0001
-
1.0000
<.0001
<.0001
0.0 (-0.0-0.0)
0.2 (0.0-0.3)
0.0 (0.0-0.0)
0.0 (0.0-0.0)
-
0.3 (0.1-0.5)
-
-0.0 (-0.0-0.0)
0.0 (0.0-0.0)
0.1 (0.0-0.1)
Meat & Alternatives
Meats / Poultry (plain, seasoned, coated)
Meats / Poultry (with sauce)
Ground Meats (plain, seasoned)
Patties, meatballs, etc.
Sausages
Deli Meats / Ham
Fish and Seafood (plain)
45
103
6
126
117
213
60
0
4
1
6
1
4
0
(0.0)
(3.9)
(16.7)
(4.8)
(0.9)
(1.9)
(0.0)
13
18
6
25
5
55
49
(28.9)
(17.5)
(100.0)
(19.8)
(4.3)
(25.8)
(81.7)
0.0002
0.0001
0.0625
<.0001
0.1250
<.0001
<.0001
-0.0 (-0.0--0.0)
0.3 (0.1-0.6)
0.0 (0.0-0.0)
0.3 (0.1-0.5)
0.3(-0.2-0.8)
0.1 (0.0-0.2)
0.0 (0.0-0.0)
132
Category N Products carrying Health Check™ N (%)
Products qualifying for Health Check™ N (%)
P-Value* Kappa Statistic (confidence interval)†
Fish and Seafood (seasoned or coated)
Fish and Seafood (sauced)
Canned Fish and Seafood (packed in broth or water)
Canned Fish and Seafood (seasoned, sauced)
Processed Fish
Dried Legumes
Frozen and Canned Legumes (plain)
Canned Legumes (prepared)
Tofu (plain)
Vegetarian Meat Alternatives
Vegetarian Terrines, Spreads or Pâtés
Eggs
Egg Substitutes
Nuts, Seeds or Ready to Eat Dried Legumes
Nuts and Seeds Butters
Legume-based dips
104
32
86
45
7
78
95
28
15
28
0
37
0
113
39
33
1
3
1
3
2
1
2
10
3
7
-
17
-
0
1
0
(1.0)
(9.4)
(1.2)
(6.7)
(28.6)
(1.3)
(2.1)
(35.7)
(20.0)
(25.0)
(-)
(46.0)
(-)
(0.0)
(2.6)
(0.0)
21
7
82
26
3
78
37
10
14
17
-
37
-
68
30
23
(18.3)
(21.9)
(95.4)
(57.8)
(42.9)
(100.0)
(39.0)
(35.7)
(93.3)
(60.7)
(-)
(100.0)
(-)
(60.2)
(76.9)
(69.7)
<.0001
0.1250
<.0001
<.0001
1.0000
<.0001
<.0001
1.0000
0.0009
0.0010
-
<.0001
-
<.0001
<.0001
<.0001
0.1 (-0.0-0.2)
0.5 (0.2-0.9)
0.0 (-0.0-0.0)
0.1 (-0.0-0.2)
0.7 (0.2-1.0)
0.0 (0.0-0.0)
0.1 (-0.0-0.2)
0.8 (0.6-1.0)
0.0 (-0.0-0.1)
0.4 (0.1-0.6)
-
-0.0 (-0.0--0.0)
-
-0.0 (-0.0--0.0)
0.0 (-0.0-0.0)
-0.0 (-0.0--0.0)
Oils & Fats
Oils
Margarines
Light Margarines
105
47
5
4
10
2
(3.8)
(21.3)
(40.0)
67
39
4
(63.8)
(82.9)
(80.0)
<.0001
<.0001
0.5000
0.0 (0.0-0.0)
0.1 (0.0-0.2)
0.3 (-0.3-0.8)
133
Category N Products carrying Health Check™ N (%)
Products qualifying for Health Check™ N (%)
P-Value* Kappa Statistic (confidence interval)†
Salad dressings 267 17 (6.4) 128 (47.9) <.0001 0.1 (0.0-0.2)
Combination Foods
Soups
Dinners and Entrees / Mixed Dishes
Pizza
Vegetarian or Meat Pies
Tofu or Meat or Fish with vegetables
Stuffed Pasta
Stuffed Meat
Major Main Entrée Sauce
Potato and Pasta Salads
Other Salads
Dried Fruit and Nut Mixture
Nut and/or Seed Bars (with or without dried fruit)
337
427
119
39
65
26
37
143
17
43
37
0
74
26
2
0
6
2
2
16
0
3
0
-
(22.0)
(6.1)
(1.7)
(0.0)
(9.2)
(7.7)
(5.4)
(11.2)
(0.0)
(7.0)
(0.0)
(-)
94
81
4
0
1
3
4
62
1
17
28
-
(27.9)
(19.0)
(3.4)
(0.0)
(1.5)
(11.5)
(10.8)
(43.4)
(5.9)
(39.5)
(75.7)
(-)
0.0005
<.0001
0.5000
-
0.1250
1.0000
0.6250
<.0001
1.0000
0.0001
<.0001
-
0.7 (0.7-0.8)
0.2 (0.1-0.3)
0.7 (.2-1.0)
-
-0.0 (-0.0-0.0)
0.8 (0.3-1.0)
0.3 (-0.2-0.8)
0.3 (0.2-0.4)
0.0 (0.0-0.0)
0.2 (-0.0-0.4)
0.0 (-0.0--0.0)
-
*P-Value for exact McNemar’s test to compare paired proportions; testing whether the proportion of products qualifying for the FOP
symbol is statistically different from the proportion carrying the FOP.
† The kappa coefficient measures the difference between observed agreement and expected agreement and lies on a scale of -1 to 1, where
0.0 is considered ‘poor’ agreement, 0.2 ‘slight’, 0.4 ‘fair’, 0.6 ‘moderate’, 0.8 ‘substantial’, and 1.0 ‘almost perfect’ agreement.
“- ” Indicates that there were no products available for a meaningful calculation.
134
Appendix B
Proportion of food products that qualified for Sens ible Solutions™ compared to the proportion of food products that
carried the systems symbol by subcategory (N=3009)
Category N Products Carrying Sensible Solutions™ N (%)
Absolute threshold criteria Based on Relative threshold criteria Products qualifying for Sensible Solutions™ N (%)
P-Value*
Kappa coefficient (confidene interval) †
Products qualifying for Sensible Solutions™ N (%)
P-Value*
Kappa coefficient†
Beverages
100% Juice‡
Refreshment Beverages
317
238
0
15
(0.0)
(6.3)
216
4
(68.1)
(1.7)
<.0001
0.0074
0.0 (0.0-0.0)
0.2 (-0.1-0.4)
-
238
(-)
(100.0)
-
<.0001
-
0.0 (0.0-0.0)
Cheese and Dairy
Natural and processed cheese
Cream cheese
Grated parmesan, Romano cheese, Cheese spreads and dip§
413
57
69
5
0
0
(1.2)
(0.0)
(0.0)
17
0
-
(4.1)
(0.0)
(-)
0.0169
-
-
-0.0 (-0.0--0.0)
-
-
404
40
64
(97.8)
(70.1)
(92.8)
<.0001
<.0001
<.0001
0.0 (-0.0-0.0)
0.0 (0.0-0.0)
-0.0 (-0.0--0.0)
Desserts 142 3 (2.1) 0 (0.0) 0.2500 -0.0 (-0.0--0.0) 103 (72.5) <.0001 0.0 (-0.0-0.0)
135
Category N Products Carrying Sensible Solutions™ N (%)
Absolute threshold criteria Based on Relative threshold criteria Products qualifying for Sensible Solutions™ N (%)
P-Value*
Kappa coefficient (confidene interval) †
Products qualifying for Sensible Solutions™ N (%)
P-Value*
Kappa coefficient†
Cookies and Crackers
536 77 (14.4) 67 (12.5) 0.3682 0.2 (0.1-0.3) 479 (89.4) <.0001 0.0 (0.0-0.0)
Salad Dressings 228 0 (0.0) 1 (0.44) 1.0000 0.0 (0.0-0.0) 215 (94.3) <.0001 0.0 (0.0-0.0)
Mayonnaise and Miracle Whip
38 3 (7.9) 0 (0.0) 0.2500 0.0 (0.0-0.0) 35 (92.1) <.0001 0.0 (-0.0-0.0)
Convenient Meals 945 13 (1.4) 432 (45.7) <.0001 0.0 (0.0-0.0) 777 (82.2) <.0001 0.0 (0.0-0.0)
Peanut Butter 26 6 (23.1) 0 (0.0) 0.0313 0.0 (0.0-0.0) 24 (92.3) <.0001 0.0 (-0.0-0.1)
*P-Value for exact McNemar’s test to compare paired proportions; testing whether the proportion of products qualifying for the FOP
symbol is statistically different from the proportion carrying the FOP.
†The kappa coefficient measures the difference between observed agreement and expected agreement and lies on a scale of -1 to 1, where
0.0 is considered ‘poor’ agreement, 0.2 ‘slight’, 0.4 ‘fair’, 0.6 ‘moderate’, 0.8 ‘substantial’, and 1.0 ‘almost perfect’ agreement.
‡Sensible Solutions™ does not have relative threshold criteria established for 100% Juice.
§ Sensible Solutions™ does not have absolute threshold criteria established for Grated parmesan, Romano cheese, Cheese spreads and dip.
“-” Indicates that there were no products available for a meaningful calculation.
136
Appendix C
Comparison of the nutrient content per reference am ount of products with and without front-of-pack nut rition rating
symbol nutrition marketing by product subcategory*
Food category Subcategory FOP status
N Calories [kcal] Saturated fat [g]
Sodium[mg] Sugar [g]
Bakery Products
Bread, excluding sweet quick, type rolls
No 147 127 [119, 135] 0.3 [0.0, 0.4] 234 [185, 264] 1 [0, 2]†
Yes 36 122 [118, 129] 0.3 [0.3, 0.5] 211 [179, 263] 2 [1, 2]
Bagels, tea biscuits, scones, rolls, buns, croissants, tortillas, soft bread sticks, soft pretzels and corn bread
No 179 147 [138, 162] 0.3 [0.2, 0.7] 266 [213, 324] 2 [1, 3]
Yes 48 146 [138, 154] 0.3 [0.1, 0.5] 217 [192, 256] 2 [1, 3]
Brownies
No 19 181 [160, 189] 2.3 [1.6, 4.2] 100 [76, 126] 15 [13, 16]
Yes 2 146 [133, 158] 1.6 [1.1, 2.1] 113 [111, 116] 15 [14, 15]
Cake [heavy weight] No 47 438 [388, 463] 12.5 [8.8, 15.0] 263 [225, 330] 33 [28, 39]
Yes 2 210 [210, 210]† 2.0 [2.0, 2.0]† 195 [190, 200] 18 [18, 18]†
Cake [medium weight] No 25 307 [288, 320] 2.7 [1.9, 10.1] 216 [107, 320] 22 [17, 36]
Yes 2 191 [145, 237]† 2.8 [2.6, 3.0] 238 [237, 238] 20 [14, 27]
Coffee cakes, donuts, Danishes, sweet rolls, sweet quick, type breads and muffins
No 82 217 [195, 236] 2.2 [1.5, 5.4] 216 [177, 249] 17 [14, 23]
Yes 7 155 [147, 173]† 0.8 [0.8, 2.0]† 165 [141, 220] 10 [9, 18]†
Cookies with or without coating or filling; graham wafers
No 231 144 [139, 155] 3.2 [1.5, 4.1] 83 [55, 110]† 10 [8, 12]
Yes 63 133 [130, 140] 1.4 [0.7, 2.0]† 104 [75, 150] 8 [7, 9]†
137
Food category Subcategory FOP status
N Calories [kcal] Saturated fat [g]
Sodium[mg] Sugar [g]
Crackers, hard bread sticks and melba toast
No 140 90 [86, 95] 0.4 [0.2, 1.3] 130 [100, 169] 1 [0, 2]
Yes 98 90 [82, 95] 0.5 [0.3, 1.1] 153 [110, 189] 1 [0, 1]
Dry breads, matzo, and rusks
No 10 110 [100, 126] 0.1 [0.0, 0.5] 177 [92, 252] 1 [0, 2]
Yes 13 130 [120, 130] 0.4 [0.3, 0.5] 170 [120, 200] 2 [1, 2]
Ice cream cones No 1 19 [19, 19] 0.0 [0.0, 0.0] 8 [8, 8] 0 [0, 0]
Yes 1 21 [21, 21] 0.0 [0.0, 0.0] 8 [8, 8] 1 [1, 1]
Croutons No 39 35 [30, 35] 0.0 [0.0, 0.1] 65 [55, 80] 0 [0, 0]
Yes 14 30 [28, 30] 0.0 [0.0, 0.5] 65 [33, 85] 0 [0, 1]
French toast, pancakes, and waffles
No 40 203 [193, 220] 1.1 [0.7, 1.6] 405 [361, 461] 6 [4, 8]
Yes 12 190 [167, 225] 0.7 [0.3, 1.6] 401 [296, 481] 6 [4, 6]
Grain, based bars with filling or partial or full coating
No 67 171 [160, 183] 2.3 [1.3, 3.2] 94 [65, 152] 14 [11, 15]
Yes 26 160 [141, 167] 1.5 [0.5, 1.9]† 98 [92, 131] 13 [9, 14]
Grain, based bars, without filling or coating
No 55 127 [118, 129] 0.5 [0.4, 0.7] 85 [69, 98] 9 [8, 10]
Yes 30 120 [111, 129] 0.5 [0.4, 1.1] 79 [60, 99] 9 [7, 10]
Rice cakes and corn cakes No 39 64 [60, 64] 0.2 [0.0, 0.3] 136 [59, 192] 1 [1, 1]
Yes 23 68 [64, 68] 0.2 [0.0, 0.3] 143 [56, 204] 1 [0, 4]
Pies, tarts, cobblers, turnovers, other pastries
No 90 331 [300, 380] 7.5 [6.1, 7.9] 204 [180, 259] 19 [17, 27]
Yes 4 158 [149, 171]† 0.0 [0.0, 0.5]† 270 [156, 270] 13 [10, 17]†
Pie crust No 12 143 [129, 150] 3.1 [2.3, 3.7] 107 [96, 142] 0 [0, 2]
Yes 1 133 [133, 133] 1.7 [1.7, 1.7] 100 [100, 100] 1.7 [1.7, 1.7]
Cereals and Hot breakfast cereals, such No 23 150 [149, 158] 0.4 [0.3, 0.4] 183 [115, 243] 12 [8, 13]
138
Food category Subcategory FOP status
N Calories [kcal] Saturated fat [g]
Sodium[mg] Sugar [g]
other grain products
as oatmeal, or cream of wheat
Yes 34 156 [150, 158] 0.4 [0.3, 0.5] 173 [111, 219] 8 [0, 11]†
Ready-to-eat breakfast cereals, puffed and coated, flaked, extruded, without fruit or nuts, very high fibre cereals
No 47 116 [111, 120] 0.0 [0.0, 0.3]† 130 [125, 200] 7 [4, 10]
Yes 38 113 [110, 120] 0.2 [0.1, 0.3] 168 [140, 214] 6 [4, 10]
Ready-to-eat breakfast cereals, fruit and nut type, granola
No 63 212 [200, 226] 0.4 [0.2, 1.4] 135 [50, 220] 11 [9, 14]
Yes 82 214 [202, 220] 0.4 [0.2, 0.6] 145 [50, 238] 11 [9, 14]
Bran and wheat germ No 1 50 [50, 50] 0.1 [0.1, 0.1] 0 [0, 0] 0 [0, 0]
Yes 1 35 [35, 35] 0.0 [0.0, 0.0] 0 [0, 0] 0 [0, 0]
Grains, such as rice or barley
No 73 160 [160, 162] 0.0 [0.0, 0.2] 0 [0, 1] 0 [0, 0]
Yes 12 165 [160, 185] 0.1 [0.0, 0.3] 3 [0, 13] 0 [0, 0]
Pastas without sauce No 322 300 [300, 307] 0.2 [0.0, 0.3]† 0 [0, 2]† 2 [1, 3]†
Yes 61 310 [310, 310] 0.3 [0.3, 0.5] 2 [0, 2] 3 [2, 3]
Stuffing No 13 167 [144, 208] 1.3 [1.3, 3.1] 557 [485, 613] 3 [2, 3]
Yes 1 167 [167, 167] 1.3 [1.3, 1.3] 359 [359, 359] 2 [2, 2]
Combination dishes
Measureable with a cup, such as casserole, hash, macaroni and cheese with or without meat, pot pie, spaghetti with sauce, stir fry, meat and poultry casserole, baked and refried beans, wieners and
No 442 316 [264, 350] 2.2 [0.8, 3.7] 820 [666, 940] 5 [2, 8]
Yes 129 275 [242, 301] 1.3 [0.6, 2.5]† 580 [528, 700]† 6 [4, 10]
139
Food category Subcategory FOP status
N Calories [kcal] Saturated fat [g]
Sodium[mg] Sugar [g]
beans, meat chilli, chilli with beans, creamed chipped beef, beef or poultry ravioli in sauce, beef stroganoff, poultry a la king, Brunswick stew, goulash, stew, ragout or poutine
Not measureable with a cup, such as burritos, egg rolls, enchiladas, pizza, pizza rolls, sausage rolls, pastry rolls, cabbage rolls, quiche, sandwiches, crackers and meat or poultry lunch-type packages, gyros, burger on a bun, frank on a bun, calzones, tacos, pockets stuffed with meat, lasagna ,chicken cordon bleu, stuffed vegetables with meat or poultry, shish kabobs, empanadas, fajitas, souvlaki, meat pie or tourtiere
No 310 306 [251, 345] 4.5 [3.1, 6.0] 644 [524, 799] 4 [2, 6]
Yes 59 195 [148, 272]† 1.9 [1.2, 2.7]† 468 [325, 558]† 4 [3, 6]
Hors d’oeuvres No 100 125 [96, 150] 2.1 [0.6, 3.8] 214 [188, 251] 1 [1, 2]†
Yes 4 88 [88, 88]† 0.1 [0.1, 0.2]† 156 [148, 177]† 2 [2, 2]
Dairy Cheese, including cream No 348 100 [86, 120] 5.0 [3.8, 6.0] 200 [160, 230] 0 [0, 1]
140
Food category Subcategory FOP status
N Calories [kcal] Saturated fat [g]
Sodium[mg] Sugar [g]
products and substitutes
cheese and cheese spread Yes 32 86 [65, 90] 3.5 [2.4, 3.6]† 210 [200, 240] 0 [0, 1]
Cottage cheese No 14 100 [100, 110] 1.0 [0.5, 1.5] 420 [305, 540] 6 [5, 6]
Yes 9 111 [100, 122] 0.6 [0.6, 0.6] 299 [277, 300]† 9 [5, 11]
Cheese used as an ingredient, such as dry cottage cheese or ricotta cheese
No 9 86 [80, 100] 4.0 [2.4, 4.7] 71 [70, 71] 2 [2, 2]
Yes 1 50 [50, 50] 1.5 [1.5, 1.5] 70 [70, 70] 2 [2, 2]
Quark, fresh cheese and fresh dairy desserts
No 62 286 [265, 300] 13.3 [11.7, 15.0] 900 [464, 1000] 0 [0, 0]
Yes 1 214 [214, 214] 7.1 [7.1, 7.1] 393 [393, 393] 4 [4, 4]
Cream and cream substitute
No 22 30 [20, 50] 1.0 [0.4, 3.0] 7.5 [5.0, 10.0] 1 [0, 3]
Yes 1 50 [50, 50] 3.0 [3.0, 3.0] 0 [0, 0] 0 [0, 0]
Milk, evaporated or condensed
No 16 20 [15, 60] 0.5 [0.2, 0.8] 16 [15, 20] 1 [1, 11]
Yes 1 65 [65, 65] 1.0 [1.0, 1.0] 18 [18, 18] 11 [11, 11]
Plant-based beverages, milk, buttermilk and milk-based drinks, such as chocolate milk
No 138 130 [100, 160] 1.0 [0.3, 2.5] 120 [100, 140] 12 [10, 19]
Yes 27 130 [110, 188] 0.6 [0.4, 2.5] 120 [110, 135] 12 [10, 29]
Yogurt No 69 140 [110, 158] 1.8 [0.1, 2.5] 93 [85, 105] 21 [9, 23]
Yes 26 125 [84, 158] 0.5 [0.0, 1.8]† 97 [88, 108] 20 [7, 23]
Fats and oils Butter, margarine, shortening, lard
No 72 70 [70, 70] 2.5 [1.5, 5.0] 60 [0, 70] 0 [0, 0]
Yes 19 70 [70, 70] 1.0 [1.0, 1.0]† 70 [60, 70] 0 [0, 0]
Vegetable oil No 97 80 [80, 80] 1.3 [1.0, 1.5] 0 [0, 0] 0 [0, 0]
141
Food category Subcategory FOP status
N Calories [kcal] Saturated fat [g]
Sodium[mg] Sugar [g]
Yes 8 80 [80, 80] 1.0 [0.8, 1.0] 0 [0, 0] 0 [0, 0]
Dressing for salad No 196 100 [60, 120] 1.0 [0.6, 2.0] 300 [240, 360] 2 [1, 4]
Yes 32 60 [30, 90]† 0.8 [0.0, 1.0] 280 [245, 320] 2 [2, 4]
Mayonnaise, sandwich spread and mayonnaise-type dressing
No 19 60 [40, 100] 0.5 [0.5, 1.0] 115 [90, 130] 1 [0, 2]
Yes 20 48 [30, 50] 0.5 [0.3, 1.0] 130 [115, 140] 1 [0, 2]
Oil, spray type No 11 4 [4, 4]† 0.0 [0.0, 0.0] 0 [0, 0] 0 [0, 0]
Yes 2 5 [5, 5] 0.0 [0.0, 0.0] 0 [0, 0] 0 [0, 0]
Fruit and fruit juices
Fruit, fresh, canned or frozen, except those listed as separate item
No 118 96 [84, 108] 0.0 [0.0, 0.0] 10 [0, 18] 18 [16, 23]
Yes 50 84 [64, 108] 0.0 [0.0, 0.0] 0 [0, 12]† 17 [12, 22]
Dried fruit, such as raisins, dates or figs
No 62 120 [110, 130] 0.0 [0.0, 0.0] 0 [0, 5] 23 [15, 26]
Yes 7 140 [140, 160] 0.0 [0.0, 0.0] 0 [0, 140] 26 [20, 26]
Juices, nectars and fruit drinks represented for use as substitutes for fruit juices
No 422 120 [110, 130] 0.0 [0.0, 0.0] 13 [5, 25]† 26 [23, 30]
Yes 132 120 [110, 140] 0.0 [0.0, 0.0] 25 [19, 35] 26 [22, 29]
Meat, poultry, their products and substitutes
Luncheon meats; pate, sandwich spread, potted meat food product; taco fillings; meat pie fillings and cretons
No 99 60 [55, 85] 0.5 [0.3, 1.3] 523 [480, 600] 1 [0, 1]
Yes 9 62 [57, 69] 0.2 [0.2, 0.3]† 314 [256, 322]† 0 [0, 1]
Sausage products No 118 148 [127, 173] 4.4 [3.0, 5.5] 528 [468, 602] 1 [0, 1]
Yes 6 85 [81, 99]† 1.3 [1.1, 1.8]† 412 [338, 587] 1 [1, 2]
142
Food category Subcategory FOP status
N Calories [kcal] Saturated fat [g]
Sodium[mg] Sugar [g]
Cuts of meat and poultry without sauce, and ready-to-cook cuts, with or without breading or batter, including marinated, tenderized and injected cuts
No 34 240 [200, 259] 3.0 [1.3, 4.0] 568 [450, 750] 0 [0, 1]
Yes 13 150 [125, 170]† 0.5 [0.4, 1.0]† 169 [106, 450]† 0 [0, 1]
Patties, cutlettes, chopettes, steakettes, meatballs, sausage meat and ground meat, with or without breading or batter
No 109 210 [160, 242] 4.2 [1.6, 7.0] 414 [311, 560] 0 [0, 1]
Yes 35 150 [120, 171]† 1.5 [0.6, 3.5]† 375 [301, 410] 1 [0, 2]
Cured meat products No 32 98 [70, 126] 2.0 [1.0, 2.9] 539 [476, 789] 1 [0, 1]
Yes 1 91 [91, 91] 0.8 [0.8, 0.8] 501 [501, 501] 0 [0, 0]
Canned meat and poultry No 25 74 [63, 85] 1.6 [1.1, 1.9] 413 [370, 487] 0 [0, 0]
Yes 6 69 [68, 70] 0.4 [0.2, 1.0]† 241 [167, 328]† 0 [0, 0]
Meat and poultry with sauce, such as meat in barbecue sauce or turkey with gravy, but excluding combination dishes
No 90 266 [210, 300] 4.2 [2.1, 5.6] 782 [605, 1092] 4 [1, 11]
Yes 16 134 [123, 145]† 0.5 [0.4, 0.6]† 442 [355, 475]† 1 [0, 2]†
Snacks Chips, pretzels, popcorn, extruded snacks, grain-based snack mixes and fruit-based snacks, such as fruit chips
No 297 250 [227, 268] 1.5 [1.0, 2.0] 330 [240, 425] 1 [0, 2]†
Yes 78 205 [176, 238] 0.7 [0.0, 1.0]† 261 [74, 360] 2 [0, 9]
Nuts or seeds for use as No 79 288 [240, 320] 3.5 [2.5, 5.0] 75 [10, 170] 3 [2, 13]
143
Food category Subcategory FOP status
N Calories [kcal] Saturated fat [g]
Sodium[mg] Sugar [g]
snacks Yes 9 311 [310, 311] 4.5 [4.4, 5.0] 156 [40, 156] 2 [2, 3]
Soups Soups No 244 90 [60, 150] † 0.5 [0.0, 1.5] 740 [650, 898] 2 [1, 4] †
Yes 90 120 [90, 150] 0.5 [0.2, 1.0] 625 [480, 650] 4 [2, 7]
Vegetables Vegetables without sauce No 250 30 [25, 50] 0.0 [0.0, 0.0] 170 [20, 270] 2 [1, 4]
Yes 84 35 [23, 50] 0.0 [0.0, 0.0] 65 [13, 290] 3 [1, 4]
Vegetables with sauce No 25 35 [21, 45] 0.0 [0.0, 0.0] 680 [486, 760] 1 [0, 4]
Yes 2 48 [35, 60] 0.0 [0.0, 0.0] 175 [0, 350] 4 [2, 6]
Lettuce and sprouts No 27 13 [10, 15] 0.0 [0.0, 0.0] 20 [13, 47] 1 [0, 1]
Yes 17 15 [11, 15] 0.0 [0.0, 0.0] 15 [11, 34] 1 [0, 2]
Vegetable juice and vegetable drink
No 25 60 [50, 60] 0.0 [0.0, 0.0] 640 [350, 640] 10 [8, 10]
Yes 16 50 [50, 60] 0.0 [0.0, 0.0] 415 [133, 565]† 8 [6, 9]
Olives No 46 20 [18, 25] 0.3 [0.2, 0.4] 237 [176, 280] 0 [0, 0]
Yes 1 17 [17, 17] 0.0 [0.0, 0.0] 131 [131, 131] 0 [0, 0]
Vegetable pastes, such as tomato paste
No 7 20 [20, 35] 0.0 [0.0, 0.0] 20 [20, 30] 3 [3, 5]
Yes 4 23 [20, 25] 0.0 [0.0, 0.0] 150 [20, 290] 3 [3, 3]
Vegetable sauce or puree, such as tomato sauce or tomato puree
No 6 24 [15, 30] 0.0 [0.0, 0.0] 200 [10, 250] 3 [2, 3]
Yes 10 29 [24, 30] 0.0 [0.0, 0.0] 200 [20, 226] 3 [3, 3]
*Front-of-pack nutrition rating symbol nutrition marketing includes: 1) nutrient-specific symbols based on claim criteria, 2) summary
indicator symbols, 3) food group information symbols, and, 4) hybrid symbols; and, excludes nutrient-specific systems that display the
amount of calories and select nutrients per serving. All data are presented as Median and Interquartile range [Q1, Q3]. Calorie and
nutrient amounts are expressed per reference amount rounded to the number of decimal places provided in the Nutrition Facts table.
144
Reference amounts are reference serving size amounts found in Schedule M of Canada’s Food and Drug Regulations and are the basis of
the criteria for making nutrient content and health claims in Canada.
† Statistically significant (p<0.05) and nutritionally relevant (≥25%) difference between products with and without a front-of-pack symbol
in the amount of calories or nutrient of interest.
145
Copyright Acknowledgements
™The Health Check logo, Health Check word mark, and the Heart and Stroke Foundation word
mark are trademarks of the Heart and Stroke Foundation of Canada used under license.